How About 2014?

SJVN, as usual, has a good piece on how FLOSS and */Linux fared in 2014. He touches on the bad things like the systemd wars and the good things like M$ loving */Linux and Android/Linux taking over the world.

I think the systemd wars will not be a brief affair. It’s taken a year or two of infighting amongst developers before it affected the much larger group of users. For some users of desktops, systemd is largely hidden and in most cases may actually make systems boot faster so it’s not an issue. In my case, I found the complexity of having to wait for every server on localhost to run before I could even log in appalling. The way Debian set it up there was just no telling systemd to give my servers the same priority as the rest of the Internet. Sigh… I can go back to Wheezy for a year or two to avoid more hassles but folks with tons of servers are in it up to their eyeballs. systemd is in their face every day. Hence the possibility of a fork of Debian. I doubt that is practical. I have faith that Debian will correct the problems sooner or later. I will wait until Jessie is released before putting in a wishlist request for more flexibility. Perhaps the issues of dependencies will be tightened up. It’s just silly to require lots of software to depend on software that lots of folks hate. I really mean hate. systemd is turning Debian GNU/Linux to look more like that other OS all the time.

As for */Linux taking over the world, I think it’s inevitable. Android/Linux seems to be working on it’s third billion users perhaps by the end of 2015. At some point there will be saturation but the diversity is amazing. I saw a young lady with a Christmas gift of a CyanogenMod Android/Linux smartphone. CyanogenMod is a customization of Android/Linux which gives users more features and some independence from Google. She’s leaving a feature-phone behind as soon as she can switch “sim” cards. Within hours she’s learned to use a bunch of features including speech-to-text (It was nearly perfect)… Strangely, at about the same time her regular notebook PC (GNU/Linux) melted down (hard drive suspected). It will be interesting to see whether she even needs to replace it. This smartphone is just so powerful. Maybe I will get one and leave Beast to serving/storing stuff.

I think, whatever happens, 2015 will be a great year for users of IT all over the world and FLOSS and */Linux will be a big part of that. For December 24, 2014, StatCounter reported some fine shares of page-views for GNU/Linux:

North Korea showed only 2.33% GNU/Linux page-views. You’d think that after what happened to SONY, they’d be more careful… 😉

Of course, Android/Linux, world-wide, scored 59% of page-views from small cheap computers. That other OS was down to 2.24%.

Merry Christmas and a happy 2015, everyone!

See ​Linux and open source 2014: It was the best of years, it was the worst of years.

About Robert Pogson

I am a retired teacher in Canada. I taught in the subject areas where I have worked for almost forty years: maths, physics, chemistry and computers. I love hunting, fishing, picking berries and mushrooms, too.
This entry was posted in technology. Bookmark the permalink.

111 Responses to How About 2014?

  1. oldfart wrote, ” I see no reason why those of us who are blessed to be able to afford a particular technology should care about either “the digital divide” or what you call “small cheap computers” but anyone else calls “cheap crappy, limited, computing” Because that’s all all your “small cheap computers” wind up being in the end – slow, limited, flimsy dead ends.”

    Leviticus 19:verse 9-10 “And when ye reap the harvest of your land, thou shalt not wholly reap the corners of thy field, neither shalt thou gather the gleanings of thy harvest.
    And thou shalt not glean thy vineyard, neither shalt thou gather every grape of thy vineyard; thou shalt leave them for the poor and stranger: I am the LORD your God.”

    We are our brother’s keeper. Strangers too. There are many reasons to encourage the use of FLOSS and small cheap computers: energy saving, reducing CO2 emissions, reducing the load on landfills, preserving our natural resources, the fact that there is no merit in having a powerful machine idling etc. I especially like small cheap computers because they are small. Smaller is better. That’s what microprocessors are all about.

  2. oldfart says:

    “So many missed opportunities followed by decades of the Digital Divide and slavery to M$.”

    What missed opportunities are these Robert Pogson? Perhaps Microsoft was in the right place at the right time, but their market was built on an entire ecosystem of vendors selling all sorts of software that people actually WANTED to use. Vendors that were able to perform tasks that were not possible before. Only someone who is foolish or blind or limited in their view of the possible refers those who are successfully making use of these programs to create and get work done as slaves.

    As far as the digital divide is concerned, I see no reason why those of us who are blessed to be able to afford a particular technology should care about either “the digital divide” or what you call “small cheap computers” but anyone else calls “cheap crappy, limited, computing” Because that’s all all your “small cheap computers” wind up being in the end – slow, limited, flimsy dead ends.

    Spare us the socialist bushwah.

  3. oiaohm says:

    Robert Pogson the same pocket case problem happened with Linus and the start of the open source BSD OS’s and if the pocket case had not happened Linux would not exist. Even by snail mail not every university was sent Unix in 1973 or 1974 it was 1976 before they were all covered.

  4. oiaohm wrote, “The timeline does not even allow for Kildall to be influenced by Unix for CP/M.”

    He was writing CP/M in the same year UNIX was delivered to academics all over the planet. Those folks may not have had Internet but they had snail mail, journals, conferences, and telephones. I know. I was there. On my campus we relied on one mainframe and a few mini-computers. When a PDP11 with UNIX showed up everyone knew it and talked about it. It wasn’t much competition in terms of throughput but it was cutting-edge technology and everyone knew it mattered. The mainframe was largely kept under glass and mostly was accessed by batch-processing. TSO etc. were added to give interactivity but it was quite limited compared to what a mini-computer could give and later the PC. UNIX was a natural for the PC and IBM and Kildall made tragic errors in letting M$ run things. Kildall might not have been any better but he had the choice and missed it. M$ definitely had the choice and did nothing but feather their bed. Apple fought UNIX for many years but finally saw the light. One wonders how different IT would have been with a UNIX OS from Day One of the PC. Too bad GNU did not start a decade earlier. So many missed opportunities followed by decades of the Digital Divide and slavery to M$.

  5. oiaohm says:

    Robert Pogson
    My point is not that Kildall copied much from UNIX but that he was influenced by it.
    The timeline does not even allow for Kildall to be influenced by Unix for CP/M.

    Kildall influenced by the same OS’s that influenced Unix developers is possible. Kildall was at one of the later Uni to take up Unix.

    The universities began cranking out folks familiar with UNIX but Kildall could not connect with them because his OS was nothing like UNIX.
    And CP/M and DOS are very close to each other. Both nothing like Unix.

    My point is not that Kildall copied much from UNIX but that he was influenced by it. Everyone was.
    The mistake here majority were influenced by Unix after 1975. But before 1975 if you knew about Unix it was that you were in some forward looking Uni that happened to hear about it. Just happens that was not where Kildall was.

  6. oiaohm wrote, “Claiming a common ancestor somewhere between Unix and CP/M is possible.”

    It’s not DNA but influence. I was rubbing shoulders with computer-science types all through those years because I used the mainframe a lot. There was an explosion of programming languages and operating systems at the time. No one thought twice about porting any software to any language or operating system. Codes were tiny then compared to now. e.g. many PDP 11’s had less than 100KB of RAM. A single person could write an OS in a year or port one.

    My point is not that Kildall copied much from UNIX but that he was influenced by it. Everyone was. The time was ripe for a good operating system. IBM had some cruft on mainframes and there were a bunch of non-Free OS out there but Kildall was interested in the PC and had an opportunity to do it right but failed. That happens. At least he was out there working for a living. It’s just too bad he didn’t use a UNIX framework and advance the PC operating system industry decades sooner. Even M$ got involved with UNIX but couldn’t figure out how to exclude everyone from that market. They could with DOS, partly because it was such crud that they could sell new licences for decades “improving” it all the while locking in developers, OEMs, retailers and users thanks to IBM’s adoption.

    CP/M could have done the same but didn’t because it was scarcely better than DOS and didn’t have the salesmen. The universities began cranking out folks familiar with UNIX but Kildall could not connect with them because his OS was nothing like UNIX. C might have been an issue, part of that, but computer science graduates of those days were familiar with many languages. There was one programmer that worked on our data-collection system who wrote the software in four languages, because, he said, certain languages were better for this that and the other… I wrote in a dozen high-level languages (Fortran IId, IVg, IVh, WatFor, WatFiv, Algol60, Algol68, COBOL!, PL/I, JCL, FOCAL, PRAXIS, Modula-2, PASCAL) over those years and a bunch of assemblers. That’s what folks did.

  7. oiaohm says:

    Robert Pogson “AT&T giving out sourcecode” this source code was in C. The reality that is not a programing language Gary Kildall used not in that time frame. Gary Kildall was never trained to code in C.

    http://en.wikipedia.org/wiki/C_%28programming_language%29 Why C was also only released to the general world 1972.

    Robert Pogson check you dates again. AT&T source code starts goes around Universities 1973 but only appears in a majority 1975/1976 and its the wrong language.

    Gary Kildall wrote CP/M fairly much solo without outside support or peer review.

    Robert Pogson think about it Unix is just all the wrong formats for Gary Kildall to having anything todo with at the time CP/M was being developed.

    Now unix command line itself is based on what old OS did. So there is a commonalities between Unix and CP/M but those commonalities are coming from old OS’s that Gary Kildall and Dennis Ritchie and Ken Thompson had access to.

    Yes the nasty reality of history Gary Kildall with PL/M and Dennis Ritchie with C both develop these languages and release the same year 1972. Also proceed to make a OS based on the language they created.

    Unix and CP/M just happen to be a very rare case independent development.

    Robert Pogson 1975 too late to effect CP/M design yes 1975 is when most universities get Unix. “Naval Postgraduate School” where Gary Kildall was teaching also only got Unix in 1975. The first papers from “Naval Postgraduate School” containing reference to Unix is 1975. CP/M command line was all fully locked down by 1974 with version 1.0. This is the big problem with the idea of CP/M being linked to Unix. There is no contact between Gary Kildall and Unix until after CP/M 1.0 is complete.

    Robert Pogson pre Internet of that time frame operating systems did not move very quickly between unis. It took 2 to 3 years for Unix to get to majority of universities.

    Robert Pogson design differences and time-line for CP/M to be linked to Unix just don’t add up. Claiming a common ancestor somewhere between Unix and CP/M is possible.

  8. oiaohm wrote, “1972 he completed his doctorate in computer science. He was studying computer science before Unix existed.”

    Nope. UNIX was started ~1969 and was in use by Bell Labs by 1969. Bell Labs started to distribute it to academia almost immediately to invite developers and build usage. It worked except for the PC.
    ” In October 1973, Dennis Ritchie and Ken Thompson drove up the Hudson Valley to the new IBM Research Center at Yorktown Heights to deliver the first UNIX paper at the Symposium on Operating System Principles.
    “It was a beautiful fall day,” Dennis remarked. Ken, who delivered the paper, told me: “The audience was several hundred. I was pretty nervous. The response was the normal, polite applause. I don’t recall any questions.”
    Ken was over-modest. The audience was quite enthusiastic. Ken and Dennis were immediately asked for copies of the new system.
    This put the AT&T lawyers in a bind: was a computer operating system part of “common carrier communications services”? Was AT&T required to distribute UNIX?
    The decision of the corporate lawyers was that Bell Labs should distribute UNIX to academic and research institutions at the cost of the media involved plus a shipping charge. Within a few months, several dozen institutions requested UNIX.”

    see Chapter 2 of the The Daemon, the Gnu and the Penguin
    By Peter H. Salus

    Kildall almost certainly did not use UNIX at university but as a guy interested in operating systems at the time and AT&T giving out sourcecode to academia he certainly would have been aware of it. He wrote CP/M the same year UNIX was released to academia so lots of his peers would have had access to it and would be talking about it. By summer 1975, it was at University of Waterloo. That’s how I heard about it. The beginnings of computer science at my university stemmed from graduates of Waterloo. When I was a graduate student, our computer science folks acquired a PDP11 and had UNIX running, that was in 1976, ISTR. It’s possible Kildall was insulated from that wave but I doubt it. Everyone in Silicon Valley knew about it and he was there. If he didn’t model CP/M after UNIX CP/M’s later development would certainly have been under pressure/influence of UNIX but, for whatever reasons, he did not head in that direction and missed the boat.

  9. dougman says:

    Windows (software, operating system) – A 64 bit OS, built upon 32 bit extensions and a graphical shell for a 16 bit patch to an 8 bit operating system originally coded for a 4 bit microprocessor, written by a 2 bit company, that can’t stand 1 bit of competition.

  10. oiaohm says:

    DrLoser
    Dave Cutler and NT and security is very much on-topic. Indeed, you, Fifi, brought it “on-topic.”
    No was not me who raised the topic of security. http://mrpogson.com/2014/12/26/how-about-2014/#comment-232913

    Notice say that I go off topic a lot. If you look closer my topic extentions are based on what Robert the owner of the site has said more often than not.

    DrLoser Number 1 NT was meant to be a Micro-kernel OS design. It is not performance cost was too high on x86.

    http://www3.sympatico.ca/n.rieck/docs/Windows-NT_is_VMS_re-implemented.html
    Little fact people are not aware of.
    We met with Dave Cutler, and he was adamant that IA32 was doomed and would we please get lost so he could target Alpha and then whatever 64-bit architecture was certain to replace IA32 by Intel.
    If Dave Cutler had his way our computers would not be x86 based. Alpha was way more suited to running proper Micro-kernels. This is the reality NT core design is designed for a ALPHA CPU hacked to work on a x86. Mind you arm is closer to Alpha than x86. So if we could get legacy application crap out way NT might be able to come good.

    The one good interview by Dave Cutler where he talks about the optimization modifications that was hosted on Microsoft.com is no longer up.

    DrLoser the highest security related OS’s are all Micro-kernels these include all the ones confirmed as mathematically secure.

    http://en.wikipedia.org/wiki/Architecture_of_Windows_NT

    Everything in the NT executive is meant to be independent processes with individual memory allocations. Lot exploits in NT would fail if this was the case DrLoser.

    So this is a case that NT design says X the implementation does Y and a stack of problems come out of it. Also that security processing was moved to user-space where debuggers can get mits on it.

    NT intergral subsystems are safe on a alpha because you have a proper memory security options. Some of the recent intel created memory security extentions at long last allow x86 todo what Alpha could. So question when will we see Windows updated to use them.

    Alpha copy of NT behaves very different to the X86 copies.

    NT design is good just it been forced the wrong cpu type. Result of using the wrong CPU type is increased exploit rate.

  11. oiaohm says:

    Robert Pogson
    http://en.wikipedia.org/wiki/Gary_Kildall

    Problem here one Kildall age. 1972 he completed his doctorate in computer science. He was studying computer science before Unix existed.

    ram there is a clue in dos and cp/m itself to its history. CDC could be its parent.

    The key difference is delete. Unix calls deleting a file remove. CDC and OS/360 in fact names deleting a file delete and removing was for destroying directory.

    Ram CP/M is command structure is a mix of items from OS before Unix.

    http://windowsitpro.com/windows-client/windows-nt-and-vms-rest-story

    Oh, and by “based off VMS,” I assume you mean “the design of NT was strongly influenced by Cutler’s experience of the design of VMS.” Because I can assure you that the two code-bases are 100% orthogonal to each other.
    Cutler and his staff who worked on VMS made the core design of NT.

    DrLoser no they are not 100 percent independent. You would call NT a reimplementation of VMS due to how much commonality there really is.

  12. DrLoser says:

    But Robert Pogson will never do that Doc. it knocks a huge hole in one of his primary arguments.

    Not at all true, ooldfart. Robert is a very graceful man.

    And, as an accredited scientist, he understands the Scientific Method.

    When your peers, and in this case Robert’s peer is oiaohm present irrefutable evidence that your conclusion is based upon wobbly premises, then as a Scientist, you withdraw what you now recognise as a mistaken and unsupportable assertion immediately.

    It is the Essence of Scientific Grace.

    Now, all we have to do is to wait for Robert to admit that oiaohm has unequivocably destroyed his axioms.

  13. oldfart says:

    “An explanation in less than ten thousand words of pointless drivel, please. Oh, and a cite wouldn’t hurt.”

    THen again I find it interesting that in all this time Lil’ Hammie hasn’t even tried to come up with google based credentials on the IBM SVC. Perhaps the poor lil’ fella has more cognition problems than we know.

  14. oldfart says:

    “When even oiaohm repeatedly dismisses your claims here, it’s a sign that you should gracefully admit defeat.”

    But Robert Pogson will never do that Doc. it knocks a huge hole in one of his primary arguments.

  15. DrLoser says:

    So where is my [request for a] demonstration of your expertise on the IBM SAN volume controller?

    That’s two cites you lack, Fifi.

    May I presume that you will now wander off into total irrelevance and refuse to acknowledge either request? After all:

    At times being a stubborn bastard and not answering off topic stuff is something else I am known for.

    I congratulate you on your stubborn metaphorical illegitimacy.

    However, that Dave Cutler thing? (And don’t imagine for a moment that you will be given a pass on anything to do with IBM or SANs or even system administration, part of the last of which you claimed experience with via IBM SAN configuration … which is temporarily off-topic.)

    Dave Cutler and NT and security is very much on-topic. Indeed, you, Fifi, brought it “on-topic.”

    An explanation in less than ten thousand words of pointless drivel, please. Oh, and a cite wouldn’t hurt.

  16. DrLoser says:

    NT from start was based off of VMS so was designed to be secure. There is a big problem. As admited by the lead author himself many alterations were done to NT that under mined its security.

    Come on, oiaohm, the entire façade of your supposed integrity depends upon this!

    Give us a cite, for God’s sake! (And be aware that I have a completely contradictory cite.)

  17. ram says:

    oiaohm said “CP/M command style in fact traces to OS/360 by IBM back in 1964 before the birth of Unix. What we call UNIX first appears in 1969 so 5 years before does CP/M style of commands appear.

    As I recall OS/360 commands were rather more cryptic than anything in CP/M. If anything CP/M commands were similar to the operating systems for Control Data (CDC) and Univac machines. I was only a small child at the time but I was already then using the “big iron” machines. Child labor was always big in IT, but they called it “education”.

  18. ram says:

    oiaohm wrote, “Dos interface is a copy off something older than Unix.”

    Not quite, but almost just as old. DOS started as a ripoff copy of CP/M. It even had CP/M error and copyright messages in it for years. In a related historical note, IBM’s OS/2 could run CP/M executables (as an undocumented “feature”). CP/M was written for the Intel 8080 “microprocessor” (today we would almost call an 8080 a nano-processor).

  19. DrLoser says:

    If he’d [Kildall] had the vision of RMS and the determination of Linus, the world would be a much happier place today.

    What, by implementing Unix on an IBM PC? Somebody tried that.

  20. DrLoser says:

    NT from start was based off of VMS so was designed to be secure. There is a big problem. As admited by the lead author himself many alterations were done to NT that under mined its security.

    So, Dave Cutler admitted that alterations to the design of VMS “undermined its security.”

    Well, we all know this, oiaohm. It would be a complete waste of your valuable time to give us a cite, wouldn’t it?

    Oh, and by “based off VMS,” I assume you mean “the design of NT was strongly influenced by Cutler’s experience of the design of VMS.” Because I can assure you that the two code-bases are 100% orthogonal to each other.

  21. DrLoser says:

    That other OS has been rewritten at least twice since the days of DOS but there are still vestiges of it kicking around.

    Technically, those would be two (or more, depending upon your view of Win9x as an OS) separate OSes, Robert. You’re still wriggling around pointlessly.

    I’ve already admitted that there are lines of code, modules even, in Win XP and beyond that have been cribbed from DOS. They’re pretty few and far between, though. You’ll most probably find them in places like the C compiler or Word/Excel — not in the kernel; not anywhere near access control and security, which as you know didn’t exist in DOS.

    Let’s face it, it’s going to be far easier to rewrite almost anything to target a 64 bit machine, if you start from an 8 bit architecture.

    “Vestiges” don’t count for much in the context of your original assertion:

    That other OS was obsolete from the beginning, being based on DOS which was a poor imitation of a UNIX OS.

    This reckless assertion presumes that significant amounts of DOS exist in the NT codebase. Not handfuls of lines; not parsing routines for the command box. Significant amounts.

    They do not, Robert. Your assertion is baseless.

    When even oiaohm repeatedly dismisses your claims here, it’s a sign that you should gracefully admit defeat.

  22. oiaohm wrote, “Dos interface is a copy off something older than Unix.”

    In computer science there is little that is older than UNIX. I used OS S/360 and there was nothing like UNIX commands there. We used JCL, FORTRAN II & IV, PL/I, and assembler for instance, in 1968. C of course was nowhere, but neither was CP/M. Kildall created a language and an OS using that language. He didn’t invent many concepts in operating systems at all. He copied the concepts widely disseminated in UNIX but unfortunately just a subset because he was on his own. Linus was much smarter to use GPL and inviting wide collaboration but that was easier with the Internet or at least e-mail and FTP being available. This sad story is just another example where “proprietary” is wrong compared to FLOSS. If he’d had the vision of RMS and the determination of Linus, the world would be a much happier place today.

  23. oldfart wrote, “our assertion that the current versions of windows are based on old DOS code is incorect”

    That other OS has been rewritten at least twice since the days of DOS but there are still vestiges of it kicking around. Just a few years ago I converted an entire school from FAT to NTFS, for instance, and vulnerabilities that were created in Lose 3.1 were still around in the time of Vista. It’s still current.

  24. oldfart says:

    So Robert Pogson, do we now admit that our assertion that the current versions of windows are based on old DOS code is incorect?

  25. oldfart says:

    “oldfart I see you throw around prior coding experince.”

    Lill’Hammie the liar and fraud is reading us the internet and attempting to pass it off as his own. Poor Lil’ Hammie, we know what you are.

    So where is my demonstration of your expertise on the IBM SAN volume controller?

    Piss off Lil’ Hammie!

  26. oiaohm says:

    CP/M itself is fairly old at 1974 by digital research history some other places say 1975.
    http://www.digitalresearch.biz/CPM.HTM
    http://en.wikipedia.org/wiki/Timeline_of_operating_systems

    1960-1980 is a very active time for OS designs.

    http://manx.classiccmp.org/details.php/1,3951 COPY written in caps appears in TSS-8 that is before Unix. Note Unix always used CP for copy. None of the command styles in CP/M are new. Where CP/M does not follow Unix look it follows the look of older OSs than it.

    Robert Pogson the problem here the command style in CP/M you get exact 1 to 1 matches with older non Unix operating systems. Yes Unix may have been around silicon valley but there were many other OS’s as well.

    Something to remember Robert most Unix related OS’s where coded in C.

    http://en.wikipedia.org/wiki/PL/M CP/M is coded in this that was created in 1972.
    But its source is PL/I from IBM in 1964. This leads you back to what you are looking for.

    CP/M command style in fact traces to OS/360 by IBM back in 1964 before the birth of Unix. What we call UNIX first appears in 1969 so 5 years before does CP/M style of commands appear.

    MS-Dos family tree does not lead back to Unix anywhere. Leads back to IBM OS designs.

    Kildall history has him using OS/360 when he was getting his qualifications.

    Robert Pogson not everything is related to unix. MS Dos not related to Unix. CP/M not related to Unix. If anything sections of Unix look like OS/360 interface so causing the commonality.

    Robert Pogson basically you are getting the relationship backwards. Techically Unix looks like Dos interface not Dos looks like Unixs because Dos interface is a copy off something older than Unix.

  27. oiaohm wrote, “The Dos commands don’t just resembled CP/M they have the exact function switches as CP/M. Little off to Unix commands that dos is just happens to be limited words for copy and so on. Unix like reducing commands to min number of letters.”

    Kildall, the authour of CP/M, was active near silicon valley when UNIX was all over computer-science schools in the area. You can bet he was influenced by UNIX. Too bad he wasn’t more into UNIX. This whole M$ nonsense could have been skipped.

  28. oiaohm says:

    oldfart I see you throw around prior coding experince.

    In fact Robort did get it wrong about NT yet oldfart is too incompent to know the facts.

    NT from start was based off of VMS so was designed to be secure. There is a big problem. As admited by the lead author himself many alterations were done to NT that under mined its security.

    Large sections of ntvdm for one. Directly sharing screen buffers with every application. Down grading of memory formal isolation rules to let win16 and win32 from 9x time frame work.

    Breach of design with win32 subsystem is about the worst.

    Original NT design says that every subsystem should sit on top of ntdll.dll no direct to kernel paths other than by ntdll.dll. For speed and optimization items like kernel.dll and other places in Windows directly syscall. Also nothing in kernel space should be dependant on win32 userspace. Why NT subsystems were meant to be removable and changable.

    NT OS design is brilliant. Implementation suxs badly.

  29. oldfart says:

    “oldfart then get your information correct and I will not join in. Don’t complain when its a case that you are being a fraud and I am pulling you up on it.”

    Lil’ Hammie still doesn’t get it. I do not debate with demonstrated liars and frauds like you. You will get no other answer than this until you prove that you actually have expertise on the IBM SAN Volume Controller.

    Piss off Lil’ Hammie!

  30. oiaohm says:

    Robert Pogson I understand confusion. Dos 2.0 in fact looks like CP/M not unix.

    http://en.wikipedia.org/wiki/CP/M
    http://www.computerhistory.org/atchm/microsoft-ms-dos-early-source-code/
    Paterson denied any wrongdoing. “I told him [Kildall] I didn’t copy anything. I just took his printed documentation and did something that did the same thing.”
    Yes Paterson the master author of DOS directly admits to implementing MS Dos following CP/M documentation.

    The Dos commands don’t just resembled CP/M they have the exact function switches as CP/M. Little off to Unix commands that dos is just happens to be limited words for copy and so on. Unix like reducing commands to min number of letters.

    Bit like how Linus implemented Linux following Posix books. Yes just like Linus added extra bugs.

    CP/M license was not cheap either.

    “8.3” filename format also comes from CP/M. Unix systems native file systems never have had the 8.3 pattern.

    http://en.wikipedia.org/wiki/Xenix
    Yes shock horror in the 1970~ Microsoft was selling a Unix before they ever made NT or dos got huge.

  31. oiaohm wrote, “Windows NT core design was based on VMS”.

    NT was not the beginning. I remember DOS 2.x. It was a single user OS and it was riddled with commands that resembled UNIX but were just a little off. Then there was that crazy file-system and the backwards slashes… Licences for UNIX operating systems were ~$1K so of course folks tried to work around that but they did it the wrong way, making one-off things instead of doing things the FLOSS way as RMS and Linus did, intending the code to be re-used and proper and more or less POSIX.

  32. oiaohm says:

    kurkosdr I agree that Xbox branding of a phone might help Microsoft but its a double sided sword.

    Remember enterprise users don’t want to appear to be game player.

    http://en.wikipedia.org/wiki/Xperia_Play People forget the playstation phone is a real item.

    Sony is attempting to use game console to gain market share with very limited success.

  33. oiaohm says:

    oldfart then get your information correct and I will not join in. Don’t complain when its a case that you are being a fraud and I am pulling you up on it.

    This is just another case if you making a false claim.

    oldfart you have claimed that I should prove myself exactly why should I have to prove my self to someone who loves overstating and getting stuff incorrect.

  34. oldfart says:

    “Why is it that when Win-Dohs has problems, people revert to use DOS commands to fix the damn thing??”

  35. oldfart says:

    Lil Hammie, you are a demonstrated liar and fraud. You are not welcome in this or any discussion that I am a part of. You can provide the requested demonstration or you can can piss off.

  36. oiaohm says:

    oldfart
    Where do you get this nonsense Robert Pogson! The last version of windows that had any DOS real code in it was I believe windows 9x (95, SE, SEII) Windows has been running a fully multitasking kernel for over 15 years now.

    Command.com and cmd.exe share some common code. This is the most in the face fragment of DOS left even in the most modern Windows.

    Also a block of dos code was deleted when 16 bit binaries were dropped.

    Next is a little horible. Some of NT BIOS interpret instruction interfacing code is based on the old dos INT 21 wrapper code this was mentioned in an old interview over how NT was constructed and seen in leaked source code.

    Yes running on DOS is wrong for a NT class OS. Containing spread out fragments of code from DOS is yes.

    oldfart historic code blocks have a habit of being around for a very long time. Please note the existence of some of the DOS code also showed up where there was a source code leak from Microsoft.

    That other OS was obsolete from the beginning, being based on DOS which was a poor imitation of a UNIX OS.
    oldfart you are right that Robert is wrong here but you have overstated. Windows NT core design was based on VMS design tainted with some DOS code for compatibility with PC BIOS and that compatibility code happens to be in kernel mode.

    DEC Alpha vms did not use a PC BIOS. So the interface between PC bios and NT is where some Dos/Windows 2.xx code end up in NT. Changing to EFI if it was completely would have got rid of more of the old fragments. Operation of NT design does not depend on the dos fragments being there. NT on Alpha for example was free of the dos junk in kernel mode.

    Mind you VMS was Unix imitation this is partly where the confusion starts NT is a OS based on a Unix imitation.

    Yet dos is not a imitation of Unix but another OS called CP/M right down to having the same executable format layout called com. Yes the 100 byte offset at the start of dos .com is to match CP/M program layout. CP/M was multi tasking. DOS is a very poor clone of CP/M.

    This is where things get funny.

    .com comes from CP/M
    Windows 16 bit binaries NE/LX used in windows on dos comes from OS/2
    PE is based of COFF format that start at Unix with some of the unique alterations that appeared in VMS plus some more alterations in NT.

    This is where Windows gets so much of a mess in places. Lets mix parts from 3 different OS styles into one and pray we don’t have issues.

    Worse is the fact that some of the Windows ABI were inverted compared to OS/2 ABIs meant todo the same thing to prevent Microsoft from being sued. Result is horible that values order changed is a cost to performance. Yes there is a reason why in some areas Linux out benches Windows. Historic attempt to be a smart ass to back stab IBM we are paying for today.

  37. dougman says:

    So you don’t know how to use Google. Its ok.. you can just Google on Bing then!

    *BING*

  38. DrLoser says:

    I don’t wish to repeat this ad infinitum, Robert, but since you have chosen not to respond, I beg leave to do it just this once:

    And I suppose an explanation of this no doubt temporary ABI/API confusion is out of the question?

    Absent a direct response, I think we can assume that the answer is: “Yes, it is out of the question.”

    In which case, are you dissimulating, prevaricating, or merely not especially knowledgeable on the topic?

  39. DrLoser says:

    Where do you get this nonsense Robert Pogson! The last version of windows that had any DOS real code in it was I believe windows 9x (95, SE, SEII) Windows has been running a fully multitasking kernel for over 15 years now.

    I imagine this little fact is far too obvious for a single one of you OS design geniuses to bother responding to?

    Sad, really.

  40. DrLoser says:

    He doesn’t know how to use Google….LOL

    And, had you taken the trouble to pass a High School Equivalency test (which, given your obvious acumen, should have been almost automatic, really), you would realise that “I dunno, just Google for it” doesn’t really count as a cite for specific information.

    What a sad loss to academia you are, Dougie. Although I’m tempted to abbreviate that to the first three and the last three words, separated by a caesura.

  41. dougman says:

    Re: DrLoser asks for citations

    He doesn’t know how to use Google….LOL

    *BING*

  42. “Stack-based buffer overflow in the animated cursor code in Microsoft Windows 2000 SP4 through Vista allows remote attackers to execute arbitrary code or cause a denial of service (persistent reboot) via a large length value in the second (or later) anih block of a RIFF .ANI, cur, or .ico file”
    see CVE-2007-0038

    others…
    “Metafile Image Code Execution and abbreviated MICE—is a security vulnerability in the way some versions of the Microsoft Windows operating system handled images in the Windows Metafile format. It permits arbitrary code to be executed on affected computers without the permission of their users. It was discovered on December 27, 2005, and the first reports of affected computers were announced within 24 hours.”

    See Wikimedia, Windows Metafile vulnerability

    “According to Microsoft, the flaw lies in the handling of TIFF image files by a graphics processing component in Windows Vista, Server 2008, Office 2003 to 2010 and Microsoft Lync. When exploited, the attacker’s code hidden in the image file executes on the target system with the same privileges as the current user.”

    see Microsoft in a TIFF over Windows, Office bug that runs code hidden in pics

    and

    “Most people know that .exe files are potentially dangerous, but that isn’t the only file extension to beware of on Windows. There are a variety of other potentially dangerous file extensions – more than you might expect.”

    So, users can be alert, but then M$ helped the malware artists by hiding file-extensions

    Sigh. It goes on for 49 million hits on Google yet DrLoser asks for citations.

  43. dougman says:

    Re: The last version of windows that had any DOS…

    Someone smoking crack I tell you.

    http://commandwindows.com/windows-8-commands.htm

    Why is it that when Win-Dohs has problems, people revert to use DOS commands to fix the damn thing??

  44. DrLoser wrote, “I suppose a cite is out of the question?”

    Yep. Just use Google and look at XP SP2. M$ forced it on us one night and shut down our computer lab at the school where I was working. We never did get the scanner to work again.

  45. DrLoser wrote, “the reason you gave up on Ubuntu in disgust, as you mentioned just the other day”.

    Well, with GNU/Linux, I’m not locked into Canonical and chose Debian GNU/Linux to escape that tendency to release beta-software. I do have some issues with where Debian is going but it seems like they will be able to correct the deficiencies of systemd just by tweaking the packages they created. By the time I go all in for systemd it should be debugged although I still dislike the structure of it with too many dependencies. OTOH, folks who are locked into M$ are in a deep hole.

  46. dougman says:

    KUKU: Only complain about released stuff please.

    LOL…who is complaining besides you? Just pointing the obvious misgivings on M$ latest Metroturd OS.

    Here are three more reason Windows 10 will fail:

    1. Windows 8 legacy – The user interface of Windows 8 was radically different from that of Windows 7 and, unfortunately for Microsoft, a large proportion of users just didn’t respond to it in a positive manner. That, along with some pretty scathing reviews, slapped Microsoft down straight away, and that will be difficult to recover

    2. More annoying tiles – What is M$ trying to say, that its consumers are blind and dumb? Placing 1″ squares all over the screen is innovation? What an atrocious mess of a lazy childish OS. Its bad when you have to install third party software to make the original software function to your liking.

    3. Free upgrade – If you followed M$ dictum, then you are in the clear. If not and you are still on Windows 7, tough luck buddy you have to pay up!

  47. oldfart says:

    “That other OS was obsolete from the beginning, being based on DOS which was a poor imitation of a UNIX OS. ”

    Where do you get this nonsense Robert Pogson! The last version of windows that had any DOS real code in it was I believe windows 9x (95, SE, SEII) Windows has been running a fully multitasking kernel for over 15 years now.

    Again, all you succeed in doing when you spout nonsense like this is make youself look like a loon.

  48. kurkosdr says:

    soundtrack = soundcard

  49. kurkosdr says:

    ” kurkosdr apologized for “Windows 10 teething troubles force users to reinstall Office Solution”. ”

    There is nothing to apologize for. Windows 10 is in (public) beta, you know what you signed up for genius.

    Even if the *beta* of Windows 10 makes your PC melt and/or makes your soundtrack play Nicki Minaj songs in a loop you still can’t complain because you signed up for a beta.

    Only complain about released stuff please.

  50. DrLoser says:

    Oh, and incidentally, Robert, isn’t this very “updates cause major issues” thing the reason you gave up on Ubuntu in disgust, as you mentioned just the other day?

    Let’s not deal in double standards, here.

  51. DrLoser says:

    Oh, and for good measure, Robert:

    That’s why so many updates prevent legitimate applications written by “partners” according to M$’s rules from running afterwards.

    I suppose a cite is out of the question?

  52. kurkosdr says:

    “Meanwhile, the ordinary user doesn’t want anything to do with fixing IT when it’s M$’s fault. ”

    Correct. MS doesn’t understand how Windows mannerisms like “failure configuring…” ruins trust in the Windows brand. Windows Phone shouldn’t have a Windows brand.

    Users can take lots of things, as Shitdroid proves, but they hate it when things fail to boot.

    I will get flamed to ashes by our DrLoser, but MS should have given their phone platform an Xbox branding, and then have the Halo and Forza team make mini versions of the games and convince some partners to do the same. They would have a fighting chance, despite being third to the game smd despite the fugly tiles.

    And tablets… Did the ability to run desktop apps on a finger-friendly device justifies the overheads, size requirements and Windows Update marathoms that come with full Windows? Sales say no.

    The whole universal apps thing should have been something between phones, tablets and xbox. Desktops Windows should be separate. I can’t imagine Adobe or Mathworks or Photoshop or any reasonably complicated desktop app becoming a universal app without shedding some significant functionality and/or efficiency anyway. The fact said companies make desktop apps proves it.

  53. DrLoser says:

    And I suppose an explanation of this no doubt temporary ABI/API confusion is out of the question?

  54. DrLoser says:

    That other OS has a very weak concept of user-space.

    I suppose a cite is out of the question?

    That’s why it likes to “execute” images.

    I suppose a cite is out of the question?

  55. DrLoser wrote, “Well, “applications being in user-space” is sort of a circular definition, isn’t it? Not very helpful, unless you’re referring to device drivers that are part kernel-space, part user-space.”

    That other OS has a very weak concept of user-space. That’s why it likes to “execute” images. That’s why attackers find it so easy to gain admin privileges. That’s why so many updates prevent legitimate applications written by “partners” according to M$’s rules from running afterwards. That’s why a whole industry sprang up to try to hold that other OS together.

  56. oldfart wrote, “with now obsolete versions of windows that were deployed on many times inadequate and/or substandard equipment”.

    That’s crap. You know it. That other OS mostly comes installed by OEMs who are M$’s “partners”. M$ spells out the specs.

    That other OS was obsolete from the beginning, being based on DOS which was a poor imitation of a UNIX OS. It wasn’t multi-user nor multi-tasking from the start, a very poor design. It had no concept of security whatsoever. Current versions treat security as a bag thrown over the crap underneath. There are huge numbers of holes found in the bag every year.

  57. DrLoser says:

    Oh, and Robert … if you truly want “simplicity” and “the right way to do IT,” you might consider a message-passing microkernel architecture such as QNX. (Or, if you prefer, the Mach kernel.)

    Simple, asynchronous, robust, and by definition “modular.”

    Run it up on a spare box some time!

  58. DrLoser says:

    The */Linux way, with applications being in user-space and only accessing the OS by a single simple ABI is the right way to do IT.

    Well, “applications being in user-space” is sort of a circular definition, isn’t it? Not very helpful, unless you’re referring to device drivers that are part kernel-space, part user-space. And not very many people would describe such things as “applications” and expect the listener to know what they mean.

    Also, you seem to have confused ABI with API. Would you care to expand on this confusion?

  59. DrLoser says:

    At times being a stubborn bastard and not answering off topic stuff is something else I am known for.

    You could have fooled the rest of us, oiaohm. “Answering off topic stuff” appears to be one of your favourite things to do.

    Actually answering a direct question? Not so favourite.

  60. oldfart says:

    “That’s poor design of software. ”

    I was not aware that you were trained as a design engineer Robert Pogson. Unless you can demonstrate that you have that experience, your comments on windows “design” remain nothing more than an uninformed biased opinion.

    There was a time when I worked extensively in various assemblers (x86/87, 8080/Z80, system 360/370 and M68000) as well as in Fortran and C. The largest project that I handled was porting the CMUSIC composition system from Unix to DOS. I can safely say that I have quite a bit more experience programming that you do. Unlike you however I know that my experience does not even begin to quality me to comment on large systems designs.

    IHMO all you succeed in doing when you pontificate on Windows design that are clearly based on experiences that you have had with now obsolete versions of windows that were deployed on many times inadequate and/or substandard equipment it so make your self look like some old kook ranting about what he can not change.

    But go right ahead Robert Pogson and continue. You do more damage to your own cause than any of us so called trolls could.

  61. oldfart says:

    “Olderman did say he would not respond to any of my posts until I did provide that data.”

    What I did say was that any time you tried to join into a discussion I was having that I would call you out for the liar and fraud that you have demonstrated that you are.

    Again, you do not get to define my terms here. You remain a demonstrated liar and fraud until you demonstrate that you have the particular expertise that you claimed.

  62. kurkosdr apologized for “Windows 10 teething troubles force users to reinstall Office Solution”.

    That’s poor design of software. Tight integration which is listed as one of that other OS’s features is bound to make updating difficult. The */Linux way, with applications being in user-space and only accessing the OS by a single simple ABI is the right way to do IT. That tight integration is killing IE as well. Now it’s the office suite being held back by the OS. Meanwhile, the ordinary user doesn’t want anything to do with fixing IT when it’s M$’s fault. That’s why Android/Linux and Chrome OS are very popular with consumers/schools. Why business thinks it’s to their benefit to mess around with M$’s OS is beyond me. One gets in trouble building one’s house on sand…

  63. kurkosdr says:

    ” Windows 10 teething troubles force users to reinstall Office Solution: ”

    Since Win10 hasn’t been officially released, I don’t think you have a right to expect quality of a finished product.

    But the “failure configuring…” nonsense in released versions of Windows is unacceptable. You buy a new computer, and you get weird errors from day1.

  64. oiaohm says:

    oldfart by how does “IBM SAN Volume Controller” relate to the current topic set by Robert.

    Please explain. If you cannot linked it to the current topic you are spaming for no good reason. Of topic if Robert bans you again I guess you will get upset. He did ban you from posting before.

    The change in handling of security reporting is a 2014 thing. The fact of running out of date drivers becoming a legal and security thing is 2014 thing.

    At times being a stubborn bastard and not answering off topic stuff is something else I am known for.

  65. oiaohm says:

    oldfart and I absolutely don’t care what you think of me.

    I am not reversing my selection not to ever answer question that was incorrectly asked. This is something I am known for. You ask the wrong way something of me I will never ever answer you on it. This is a matter of my honor code.

    Reality oldfart is a Grey beard Zealot.

  66. oldfart says:

    “Since Lil’Hammie is not officially one of my handles ”

    You don’t get it Lil’ Hammie,. You have been given your to-do. You can fuss and fume and call me whatever you wish. You remain in my eyes a proven liar and fraud who has no right to be called anything else.

    Should you be able to prove that you know about the IBM SAN Volume Controller, then I will return to calling you by your chosen nym and will return to debating you.

    Until then, anything else you say to or about me means nothing more than the dribblings of a little child.

  67. oiaohm says:

    Going forwards there is big trouble. Most security groups are now taking the mandatory 90 release after informing vendor or security flaws. Because it turns out with holding the fact security flaw exists can get you sued.

    http://threatpost.com/eff-sues-nsa-odni-over-zero-day-disclosure-processes/106971
    Yes the NSA was one of the first sued over stock piling security flaws.

    Current hearings have gone in the eff and other parties attacking over the withheld information favor. The means to pay for security flaws to be hidden for years on end may be over for good and anyone taking part in this in future could have ass sued off badly.

    How long until using unmaintained software becomes a legal problem who knows.

  68. oiaohm says:

    Sorry messing around with fake ids expect to be bundled in with DrZealot.

    Olderman did say he would not respond to any of my posts until I did provide that data. In fact asking for the data repeating is breaking exactly what you promised. I do have the exact post about time you find it and go back and read what you promised todo. Once you made that statement I had no reason ever to provide you with the answer because every time since you asked is proof you cannot keep your word.

    Only a Zealot who is alters someone name.

    Piss off Lil’Hammie!
    Since Lil’Hammie is not officially one of my handles I don’t have to respect this request at all. I really should call you MrZealot due to your Zealot tendencies.

  69. dougman says:

    Yes, M$ version of reliable IT.

    “Failure configuring Windows updates. Reverting changes. Do not turn off your computer”

    Here is another example.

    Problem: Windows 10 teething troubles force users to reinstall Office Solution: Reboot, Uninstall Office, Reboot, Install the hotfix, Reboot, Reinstall Office, Reboot

    One more: http://www.engadget.com/2015/01/02/google-posts-unpatched-microsoft-bug/

    Its hilarious reading the comments whining over “how dare Google do this to M$”…LOL
    https://code.google.com/p/google-security-research/issues/detail?id=118

  70. oldfart says:

    “oldfart o you are DrZealot”

    And Lil’ Hammie demonstrates that he is a liar and a fraud again. This time to Robert Pogson, who knows that I am who I am.

    Piss off Lil’Hammie!

  71. oldfart says:

    “Funny oldfart is a security idiot.”

    And Lil’Hammie is demonstrated liar and a fraud who is unwelcome in this discussion. You have yet to demonstrate that you ever had any knowledge of the workings of the IBM SAN Volume Controller. You have also lied when you stated that Olderman “Promised” not to talk to you – I made no such promise to you.

    So Piss off Lil’ Hammie!

  72. oiaohm says:

    oldfart o you are DrZealot

  73. oldfart says:

    Oh, and I believe that enough doubts have been cast on your interpretation of statistics in previous discussions as to render them of dubious utility.

  74. oiaohm says:

    Funny oldfart is a security idiot. Just because something is a Dell does not mean it will have maintained drivers over the life span of the computer.

    Security is a huge Elephant in the room.

    Actually Mr. K. you need to get a dell with windows 8.1 with carefully chosen hardware with vetted and tuned drivers installed for you.
    Dell machines are only vetted to have working drivers from the day you buy it. Any issues appearing after that stiff.

    Really people complain about lack of means to update phones yet they fail to see the issues with general PCs.

    Even reverting to older version of Debian does not mean having to use unmaintained.

  75. oldfart says:

    ” Provide reliable IT when that other OS was folding under its own weight and the malware it attracted?”

    Before I can answer this Robert Pogson, we need to agree on a definition of “reliable IT” I am particularly interested in such a definition the context of the CURRENT state of the art for a personal desktop OS – say, a current version of Linux vs. windows 8.1. However since I know that you have had to revert from the current version of Linux due to your issues with systemd we can discus the older version vs. ) windows 7.

    Over to You Robert Pogson.

  76. oldfart wrote, “the abortion known as Desktop linux in the trash where it has always belonged.”

    That’s rather harsh. What has desktop GNU/Linux done to deserve that? Provide reliable IT when that other OS was folding under its own weight and the malware it attracted?

    I notice that in Uruguay, on Christmas Day, GNU/Linux was getting 10% of page-views, so that’s likely in folks’ homes, not at school and not at work. On New Year’s day, it was 11.68%. Clearly, Uruguayans don’t share old fart’s pessimism and give/receive GNU/Linux as expressions of love. In Norway, page-views peaked at 5.4% on December 31. Both of these countries introduce students to GNU/Linux and it seems they are using it at home now too, of their free will. Clearly, old fart is out to lunch in his opinion of GNU/Linux on the desktop.

    In Venezuela, GNU/Linux page-views were neck-and-neck with Lose 8.1 over the Christmas to New Year’s period. Does Lose 8.1 belong in the trash?

  77. oiaohm wrote, “really the arguement about driver compatibility comes down to a security vs usability argument.”

    That’s a huge point. I’ve seen a lot of ancient drivers for that other OS on manufacturers’ sites. It’s hard to imagine them all being perfect. Things like buffer overruns and use after freeing don’t take a holiday. A remote attacker might have a hard time accessing them directly, but with a little help from a local user… Perhaps building custom kernels with a minimum of drivers all built from source code is not such a waste of time after all. I stopped doing that for systems with Nvidia drivers because they are such a pain but Beast gets my TLC.

  78. dougman wrote, “Seems Windows 8 market share is down, and someone edited the data”

    Yeah. That’s really weird. It’s interesting that both StatCounter and that other site agree to within 3 digits the share for GNU/Linux but disagree in the second digit for that other OS. That was some “correction”. How can a drop in 8.* be explained? A Christmas retail crash combined with a roaring business in 8.* to XP migrations? The SONY hackers DOSing NetApplications? M$’s astroturfers on vacation? M$ was late with their payment? Chinese firewall went down…

  79. oldfart says:

    “Your computer must be really old, save some money, use Linux and get yourself a new Dell computer.”

    Actually Mr. K. you need to get a dell with windows 8.1 with carefully chosen hardware with vetted and tuned drivers installed for you. You can then fill it with all of the FOSS, freeware or if you need it commercial for-pay software that you need.

    And leave the abortion known as Desktop linux in the trash where it has always belonged.

    Oh and Dougie, since you seem to have the notion that old = a dinosaur, I figured that I would change my nym for the new year.

    Have a nice day!!

  80. oiaohm says:

    kurkosdr
    It’s not about drivers. It’s about all those USB devices whose mamufacturer release some driver releases for some short time and then forget the device even existed.
    And its two of those drivers that was used in the sony case to install the software into the computer harddrive to delete on reboot.
    https://www.us-cert.gov/ncas/alerts/TA14-353A
    Name: usbdrv3_32bit.sys
    Size: 24280 bytes (24.3 KB)
    MD5: 6AEAC618E29980B69721158044C2E544
    PE Compile Time: 2009-08-21 06:05:32
    for example

    Instead, any 2008 driver works in Win8.1
    This is why Sony is so badly damaged is that a flawed 2009 driver would still run today so allowing the boot sectors to be overwritten without correct approval.

    Driver backwards compatibility comes at a huge price.

    Linux attitude to binary drivers is not being Lax at all. Microsoft is the one being lax where is the black list on Windows to prevent security flawed drivers from loading.

    kurkosdr really the arguement about driver compatibility comes down to a security vs usability argument. If you value security higher drivers have about an operational life of 2 years.

    When the manufacture forgets a device exists and no third party is providing updated drivers you should drop the device in bin as its is now a security risk. If you don’t do this you are not running secure computers.

  81. dougman says:

    But kuku, dear God. You never answered my question and I will summarize it here for you.

    Why should I or anyone else use a driver that is six-years old?

    Your computer must be really old, save some money, use Linux and get yourself a new Dell computer.

  82. kurkosdr says:

    It’s not about drivers = It’s not about video drivers

  83. kurkosdr says:

    It’s not about drivers. It’s about all those USB devices whose mamufacturer release some driver releases for some short time and then forget the device even existed.

    Linux Desktop’s lax attitude to driver bacj compat clashes badly with this. Instead, any 2008 driver works in Win8.1

  84. oiaohm says:

    dougman you missed that current driver is 96.43.23
    http://www.nvidia.com/object/linux-display-amd64-96.43.23-driver.html
    Yes revised due to security issues in 2012. Running 96.43.07 from 2008 is asking to be hacked. There is no safe to use video driver for Linux older than the year 2012.

    The fact Linux cannot keep drivers around for a long time due to security flaws that get found and in the sony case were an old windows driver allowed exploit(this is not uncommon).

    Security vs Backwards compatibility. Security truly does draw a line on how much backward compatibility you can have. From 2004 to 2012 is one driver time frame for Linux kernel. Yes video drivers from 2004 will install on 2012 Linux distributions. The Linux defined kernel API does not break that much just sometimes its impossible to avoid. Linux kernel undefined internal Kernel API change all the time.

  85. dougman says:

    http://www.nvidia.com/object/linux_display_x86_96.43.07.html

    Version: 96.43.07
    Operating System: Linux x86
    Release Date: July 16, 2008

    But the question would be, why 2008? Are you trying to say that you are using outdated software? Honestly, your question is stupid.

  86. oiaohm says:

    kurkosdr
    Can i load a proprietary driver or app from 2008 and have it run on the latest Linux Mint problem-free? No?
    Those are two separate questions.

    Proprietary driver from 2008 if you are referring to Nvidia and AMD video drivers the answer is no because particular X11 ABI interfaces are no more due to security flaws. This has been on going problem. Insecure driver in kernel space breached the security of the complete OS.

    Proprietary application like Maya from 2008 does run on the latest Linux Mint. It was shipped correctly with runtime bundled.

    Problem here applications shipping correctly for Long term usage are a rarity for Linux.

  87. kurkosdr says:

    “LOL…maybe ‘yours’ does, but this dudes computer (Linux Mint)”

    Can i load a proprietary driver or app from 2008 and have it run on the latest Linux Mint problem-free? No?

    Desktop Linux works fine only if you are fully into open source.

  88. dougman says:

    Interesting read: http://linuxmigrante.blogspot.com/2015/01/december-was-disaster-for-windows-8xwhy.html

    Seems Windows 8 market share is down, and someone edited the data…LOL.

  89. dougman says:

    ” before I can do the 8.1 upgrade, I have to install all 112 updates. This was going just fine until after it rebooted, and now it’s been stuck on the “Failure configuring Windows updates, Reverting changes” screen for about the past 12+ hours.”

    Wow..I think 112 updates would take me 5-minutes, perhaps even less and I know for a fact a “reboot” would not be needed at anytime. I would still be able to work, etc.. The only time a reboot would be required is if I was updating the kernel itself.

  90. dougman says:

    Re: Sometimes your PC will get stuck at an endless boot-restart-boot sequence.

    LOL…maybe ‘yours’ does, but this dudes computer (Linux Mint) /Laptop (ChromeOS) /Tablet (Android) /Smartphone (Android) /NAS (Slackware) have no such troubles.

    See the correlating item between all those devices?

  91. kurkosdr says:

    You know Bing sucks because when you type “get firefox”, it lists an unofficial site as it’s first organic result and not a mozilla site.

    “On a different note, lets talk about “failure to configure Windows updates.”…seems lots of lots of trouble for Win-Dohs users out there these days.”

    Sometimes your PC will get stuck at an endless boot-restart-boot sequence.

    http://www.tomshardware.co.uk/answers/id-2160980/windows-tablet-stuck-failure-configuring-windows-updates-reverting.html

    Share And Enjoy!

  92. dougman says:

    LOL…

    Who uses the Library of Congress?? Obviously Congress does, which consequently is opposite of progress, but I digress. As a upstanding intellectual pompous nimby you make yourself out to be, you should have known that.

    On a different note, lets talk about “failure to configure Windows updates.”…seems lots of lots of trouble for Win-Dohs users out there these days.

  93. oiaohm says:

    DrZealot
    10% URLs (yes, honestly. People type urls, or partial urls, into the query box)
    This is not understanding end user actions.

    End user types part or slightly incorrect into Internet explorers addressbar and then Internet explorer looks it up in bing. So no binging/googling URL’s is not common-ally user action but its browser action.

    Bing over view numbers are most likely wrong. DrZealot forgets web browsers has a item called a bookmark. Commonalty visited sites will be bookmarked and users will not google or bing for them. Weather forcasts would fall under item to be commonalty book marked.

    Also government weather sites don’t include advertising so are not picked up by stats counter either.

    Looking at local web cache is quite a good metric to see what your local users do use. The web cache is not effected by lack of adversing or bookmarks or client filtering software.

    Google does it. Bing does it. Lots of people do it.
    Of course here is DrZealot being is normal Zealot self. Do any of those people collect the information in a calculable error value. The answer is no they don’t.

    Reality we have very little understanding on what users are using. The best people are ISP’s. Why is Google building there own ISP that is right to correct the Google search engine. The likely wanted site on a topic is most likely the one the most people visit currently no method gives Google that information correctly. ISP/web cache data is more correct but will still have errors caused by other caches.

    The reality search engines will have to come ISP’s to be able to collect the data they need to provide good results. So far I have not seen Microsoft state when bing is going to have its own associated ISP service.

  94. DrLoser says:

    Who the hell uses Bing. When was the last time you heard these phrases??

    You got me there, Dougie. And also, who the hell uses the Library of Congress? Or even bothers to read a book every now and again?

    How’s the endless search for an HSE going, btw?

    Thing is, I’m not recommending Bing here. I am simply — and I will spell this out to you in small words, Dougie — pointing out that there are better ways to collect numbers on Web Page usage than a random squiz at a school web-cache or twenty.

    Google does it. Bing does it. Lots of people do it.

    It would be beyond your ken, I think, Dougie, but keep bashing those rocks together!

  95. dougman says:

    Who the hell uses Bing. When was the last time you heard these phrases??

    “Dude, lets Bing the football scores last night”

    “Mom, I need to use the computer for abit and Bing Lady Gaga concert dates!”

    NEVER..

    No one uses Bing for Internet searches. Hell even the now the search engine is down…

    “Bing services aren’t available right now
    We’re working to restore all services as soon as possible.
    We know you want to get back to searching. Please check back soon.”

    Bing is about as annoying as Ned Ryerson!

    https://www.youtube.com/watch?v=4wpd67fF5DQ

  96. DrLoser says:

    My experience with Bing brings up another interesting point. I worked with Related Search. The point of Related Search (this is true of Google, also) is to present the user with ~6-8 … er … Related Searches.

    Wanna know how we professionals do this? We churn petabytes of browser histories through a sophisticated machine-learning model, using tens of thousands of top-end 32-core servers, relying on the best locality info we can find, and filtering the results through a) popularity and b) something like a hundred “signals,” representing user intent.

    Your eyes glazed over, I’m sure. But I’d like you to take one thought away from all that. If “school usage” or “educational usage” was actually a “signal,” we’d have noticed it! (Why? Because M$ is Evil. Don’t ask silly questions.)

    And we didn’t.

    Pity we didn’t just ask a Provincial Canadian Government employee to grovel around in a couple of dozen web-caches. That would have been so much more convincing.

  97. DrLoser says:

    Since I worked in government for years and inspected web-caches, I do. In one place, the top site was regional weather followed closely by governmental education sites and other educational resources, none of which were clients of Statcounter.

    Anecdotal, Robert, anecdotal. Not only that, but you narrowed your focus down to “one place.” Not especially convincing.

    Now, from a slightly more broad-ranging (ie Global) perspective, yet still admittedly anecdotal, I’m going to present you with a rough breakdown of stats from Bing, which I will use as a proxy for a web-cache:

    12% Porn
    10% URLs (yes, honestly. People type urls, or partial urls, into the query box)
    10% News (a broad range, I admit)
    10% Sport
    … and 58% “something else.” From what I recall, Weather was also about 5%.

    Your imputation that school usage and/or government usage is radically different from “typical” usage is an extraordinary claim, unsupported by anything other than anecdote. (Incidentally, I’m assuming that your anecdotal “working in government” is coterminous with your anecdotal “working in education.” I shall be happy to be proven anecdotally incorrect.)

    I’m sure there are figures out there, but you haven’t really put any effort into finding them, have you?

  98. kurkosdr says:

    ” That other OS was down to 2.24%.”

    When it comes to Windows Phone, calling it “that other OS” is rather fitting.

    Congrats Ballmer (with your slow response to the iPhone) and Sinofsky (with your stupid Metro tiles) you ‘ve managed to make Pog’s newspeak sound almost sane.

  99. ram wrote, “The most damaging thing to Debian was Gnome3 being made incompatible with Gnome2.”

    Yes, those changes made no sense to me. Instead of overwriting GNOME2 they should have forked. Why change something that’s not broken? We don’t shoot old people just because they are not new… the same should apply to the working environment of millions of people. It was short-sighted and wrong. Then KDE and Canonical copied that mistake. You can introduce all kinds of good ideas without overthrowing old, well-established ideas. The pulseaudio and systemd things could be similar mistakes. One throws on another unnecessary layer and the other showers us with needless complexity. In physics, there is a principle that the simplest plausible solution is probably right. If you need wheels within wheels to describe something, you’re deluded. If there’s something lacking in the world of FLOSS it’s the idea that software should be allowed to mature rather than being constantly changed, BUGgered and replaced. There is a natural life-cycle to software and it doesn’t end with a chop. It ends with a phase-out as numbers of users and the usefulness of the software declined. There was no decline in usage of GNOME 2. Why jerk around the users?

  100. ram says:

    The most damaging thing to Debian was Gnome3 being made incompatible with Gnome2. Yes, there is Ubuntu, and then Mint with Mate that on the desktop feels like an improved Gnome2, but Mate has not (yet) been ported to earlier Linux distributions such as Debian Squeeze.

  101. oiaohm says:

    1) In what possible way could that be even remotely important?
    2) Your Google-Fu let you down. XP went EOS on 30th June 2008.
    3) You were presumably misled by something like this:

    DrZealot the moron using google strikes again. Number one I am Australian. Due to Australia Fair-trading ruling the System Builder Date applies to Retail. So the EOS on 30th of June not a value that applies in the market where I am.

    “System Builder OEMs will still be able to purchase all Windows XP versions through January 31, 2009…”
    Yes January 31, 2009 is the EOS of Australia. EOS dates for products are not identical country to country. Yes country laws effect the result.

    Considering that Windows 7 overtook Windows XP for the first time in October 2011, oiaohm, I would gently suggest that you are full of it on this one.
    DrZealot the Moron strikes again. That change lines up with Charities getting free volume license upgrades from XP to 7 in 2011. So old machines being updated. So a good percentage of 7 is 5 years or older.

    Statcounter does not tell you if the computers are new. DrZealot using numbers without understanding them.

  102. oiaohm says:

    Robert Pogson don’t worry DrZealot is being his normal moron without a brain who does not know when to keep mouth shut.

    Of course DrZealot would not bother reading this recently
    https://www.us-cert.gov/ncas/alerts/TA14-353A
    to make sure he was up on the topic.

    Section “Strategic Mitigations” item
    Implement a Secure Socket Layer (SSL) inspection capability to inspect both ingress and egress encrypted network traffic for potential malicious activity.
    In idiot terms as required for DrZealot. This basically means install Man in Middle Proxy solutions.

    Now spend some time and follow back where this recommendation comes from.

    This has been a general recommendation for years. For high quality of filtering against porn in school networks is forbidding any traffic you servers cannot decode and inspect.

    Any high security company or government or school will have proxies filtering all incoming and outgoing traffic so screwing all web stats. If they don’t they have in fact disobeyed http://www.us-cert.gov recommendations are are mal-configured networks.

    Less than 50 percent of most government computers will appear in web stats. That is just the way the cookie crumbles.

  103. DrLoser wrote, “Do you have even the vaguest statistics on “government offices”/”schools” web-surfing? I strongly suspect that you don’t.”

    Since I worked in government for years and inspected web-caches, I do. In one place, the top site was regional weather followed closely by governmental education sites and other educational resources, none of which were clients of Statcounter.

  104. DrLoser says:

    They [M$] have a good share everywhere but we know GNU/Linux is still spotty and used in government offices not likely to be surfing the web all day long (low page-views) and schools (ditto).

    Time to call you out on this one, Robert. Do you have even the vaguest statistics on “government offices”/”schools” web-surfing? I strongly suspect that you don’t.

    In fact, I believe the assertion is prima facie risible.

    If you’re a bureaucrat, there’s a good chance you’re looking things up on the Web every day. And if you’re a student with access to the Internet, I would like to think that you are looking things up on the Web every day. In both cases, far more frequently than John Doe.

    Which is of course Good for Linux, because Linux has a better foot-hold on government offices and schools (cf Reunion, Uruguay, the French Gendarmerie, whatever) than they do anywhere else.

    Which would be a win for Linux. Except, not, because these people according to you don’t use the Web. So it’s a loss for Linux.

    Sometimes, Robert, I wonder whether you remember the good old days, when you were a student Physicist and believed in the power of numbers, not contradictory fantasies.

  105. DrLoser says:

    Something important …

    … here comes another brain-fart from the expert on “importance” …

    … about that 15 percent XP is remembering XP went end of sales in 2009.

    Brain-fart right on schedule, oiaohm. Well done, little lad.

    1) In what possible way could that be even remotely important?
    2) Your Google-Fu let you down. XP went EOS on 30th June 2008.
    3) You were presumably misled by something like this:

    “System Builder OEMs will still be able to purchase all Windows XP versions through January 31, 2009…”

    … found here. Not that it matters, except that it demonstrates your complete inability to understand what you read. So, nothing new there.

    So every XP showing up should be 5 years old or older.

    By Dah Rules, yes. Unfortunately for your proposition, and even though nobody at all has a reason to care, you’re obviously wrong.

    1) Any OEM builder in Jan 2009 would pretty much certainly have inventory left in Jan 2010. Which would be XP inventory.
    2) Any corporation — hell, any top-end SME — with a Corporate License for XP might well have been installing XP right up to, and possibly beyond, the end of life deadline. Which isn’t even the end of sales deadline. For details, cf Robert’s comments on the NHS in England.

    Even a good percentage of Windows 7 machines will be 5 years old or older.

    Considering that Windows 7 overtook Windows XP for the first time in October 2011, oiaohm, I would gently suggest that you are full of it on this one.

    Have you tried a Barium enema?

  106. oiaohm says:

    Something important about that 15 percent XP is remembering XP went end of sales in 2009. So every XP showing up should be 5 years old or older. Even a good percentage of Windows 7 machines will be 5 years old or older.

  107. gamer88 wrote, “Windows still count for 89.38% of the desktop market.”

    From the desktop-os-ww-daily link:

    • “7” – 53.71%
    • “8” – 4.69%
    • “8.1” – 12.69%
    • XP – 15.22%
    • Vista – 3.07%
    • Total due to M$ – 89.38%

    We’ve discussed this many times here. The numbers are known unreliable to use this way. e.g. Apple publishes actual units shipped at around 6%, so unless Macs magically last longer than stuff made for other PCs in the same factories… they don’t have 7.74% share. You could argue that cheating all goes to M$ but I could also argue that GNU/Linux is undercounted, e.g. NAT and web-caches in schools where I’ve worked, etc. Servers don’t know every PC that loads their pages. Since M$’s stuff is widely on retail shelves, that doesn’t affect their numbers. They have a good share everywhere but we know GNU/Linux is still spotty and used in government offices not likely to be surfing the web all day long (low page-views) and schools (ditto). Wikipedia reports MacOS = 5.55%, for instance and that other OS, only 42.86% of all traffic when non-mobile gets 62.4% of traffic, so that other OS gets 42.86/62.4 X 100 = 68% of non-mobile traffic. You can argue traffic .NE. installed-base but over time they will match. M$ is definitely in decline. You can look at their SEC filings to see that. They are diversifying away from desktop clients as fast as they can because they are tanking. Folks are using smartphones and tablets and thin clients to do what Wintel PCs used to do and PCs are lasting longer…

  108. gamer88 says:

    Of course, Android/Linux, world-wide, scored 59% of page-views from small cheap computers. That other OS was down to 2.24%.

    You counted only mobile(smartphone) OSs, not desktops. And you mean just Android , not Android/(Desktop)Linux.

    http://gs.statcounter.com/#desktop-os-ww-daily-20141224-20141224-bar
    Windows still count for 89.38% of the desktop market. And for the heck of it, I lumped the two categories together(along with tablets) and windows still show 54.21%.http://gs.statcounter.com/#desktop+mobile+tablet-os-ww-daily-20141224-20141224-bar

  109. DrLoser says:

    I may be a crusty old curmudgeon, Robert, but sadly, no matter how much “saturation” there might be in Cyanogen-Mods posted over the Internet …
    … and, let’s face it, with the Billyuns and Billyuns of AT&T/C/MIT/Gnu/Linux/Android/AppStore devices out there, that is a whole lotta market opportunity goin’ on!

    … It’ll all end in tears. Just one young lady on You-Tube, or whatever, and about 10^5 people who don’t really “get” it.

    But of course, Robert, you “get” it. You not only write these leads: you research them.

    How much fun was it for you to mod your Android phone with FLOSSy Freedom, just out of interest?

  110. DrLoser says:

    Almost all the “high tech” used in Australian RSL’s originates from North Korea — talk about ironic!

    Correct me if I am wrong, ram, but assuming you are correct, that “high tech” would come from the Kaesong Industrial Complex (KIC). Which is, in tech terms, South Korean.

    Either way, it’s interesting in a sort of “noodling away the hours until death” way.

  111. ram says:

    I’m surprised North Korea showed only 2.33% GNU/Linux page-views. Considering they manufacture a huge amount of embedded Linux surveilance systems for the casino/gambling industry. Almost all the “high tech” used in Australian RSL’s originates from North Korea — talk about ironic!

Leave a Reply