The Cost of Lock-in

Repeatedly, I am told that it costs less to be locked-in to M$’s way of doing things in IT. The explanations are diverse and wonderful:

  • malware doesn’t happen,
  • M$’s stuff is just better and easier to maintain,
  • more/better applications are available for that other OS,
  • GNU/Linux is bad, the GPL is bad, no one makes anything good for GNU/Linux and
  • users want M$ so you are better off making them happy…

I think that’s all nonsense. Doing anything in IT with M$ underneath puts toll-booths on everything and prevents using the best technology to do the job. Apparently, the world is getting the message and adopting FLOSS but you could not tell that from M$’s quarterly report yesterday. They beat the market… However, M$ has bothered to plaster the government of the United Kingdom with e-mails about how it costs less to use both FLOSS and non-FREE software. Apparently governments in Europe use enough FLOSS that M$ has noticed and has given up trying to explain that they are cheaper but that a combination is cheaper. Give me a break. As Glynn Moody observes, if FLOSS-only is more expensive than mixed, so is M$-only…

That’s the trouble with FUD. It’s elastic. If M$ does not get the effect it wants, it changes its FUD. If you want the truth, use FLOSS for the OS and the office suite and notice

  • fewer re-re-reboots,
  • less corruption of systems by malware or whatever,
  • better performance because there’s just less software getting in the way of what you want to do,
  • more flexibility because the GPL and other Free Software licences give you rights instead of taking them away,
  • faster implementation of systems because you can get the necessary licences for FLOSS as fast as you can acquire the software over the Internet (the licence is provided in the software), and
  • you don’t need to constantly convert file formats and if you wish to do that, it’s easy.

With FLOSS you’ll never have to buy a licence to open a file you make. You won’t need three times as many IT people because FLOSS breaks less often. With FLOSS you’ll never have to buy new hardware because M$ and hardware makers have made a secret deal.

I recommend Debian GNU/Linux because the number and variety of software packages gives you one-stop shopping (for $0) for a complete IT system. Whether you have one PC or thousands APT packaging system makes it the same problem to update or change one or many. Tell me how that can cost more than paying someone to count licences and keeping track for lawyers and accountants of every bit of software. Tell me how Oracle v Google or M$ v TomTom or Apple v Samsung made the world a better place. Tell me how M$ made more money selling software licences last quarter than all the PC makers earned. HP earned only $464 million for its Personal Systems Group selling 18% of the world’s PCs. M$ made twice as much as all the PC OEMs put together selling permission to use PCs.

See Glynn Moody – Does Microsoft Office Lock-in Cost the UK Government £500 Million? – Open Enterprise.

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.

31 Responses to The Cost of Lock-in

  1. ch wrote, “most of the successful FLOSS stuff is either essentially a copy of something commercial (like Linux being a copy of Unix) or started life as a commercial product (like OOo/LO).”

    Apache, ImageMagick, SOX, FFMpeg, vlc, are not copies of anything. People set a goal and achieved it.

    Linux is not a copy of a UNIX operating system. It is a UNIX operating system but not certified. Building to a standard is not copying.

    FLOSS is a state of being, not as well as a method of creation. FLOSS can be created or destroyed by its authour and the software can be licensed any way the authour feels fit.

    StarOffice was already being distributed for $0 when SUN bought the company for $73million, less than the cost of one round of licences and notebooks for its 43K employees. SUN freed the source code a couple of years later. For less money they probably could have written OpenOffice.org from scratch. The actual process by which FLOSS is created is much less significant than that it is FLOSS in the end. Commerce is not an essential facet of FLOSS, but useful. Commerce provide funds to pay programmers more directly and salesmen to distribute the stuff but a lot can be accomplished without that. GNU/Linux was on millions of machines before IBM became involved.

  2. Ch says:

    Mr Pogson,

    could you please come to the rescue of my last post?

  3. Ch says:

    > The point still remains that FLOSS is not always following/copying the lead of non-FREE software.

    No, not _always_. However, most of the successful FLOSS stuff is either essentially a copy of something commercial (like Linux being a copy of Unix) or started life as a commercial product (like OOo/LO). And when FLOSS tries to be innovative, the result may well be something like Unity.

    > For many purposes one is not better than the other but when it comes to price/performance, and freedom, FLOSS is a clear winner.

    I would formulate it somewhat different:
    When it comes to features and performance, MSO is the clear winner. If that is not important for your purposes, you can use the freebee instead.

    > menus/icons are reasonable and workable

    Up to a point. Unfortunately, MSO has surpassed that point some time ago, at which point the “classical” menu just broke – not in the way of “software doesn’t work”, but in “users can’t find the new features we put into the new version”. Retaining the old menu simply wasn’t an option any more. (Check out “The Story of the Ribbon” on youtube.)

    And yes, the ribbon is actually a good solution to this – you really should give it a try ;-)

  4. Ch says:

    Mr oia…,

    > 1985-1990 Starwriter is a mix of applications in one a suite.

    No, no and no. It was just a word processor, hence the name, and not the least bit a suite – that’s what happened later: Star Office.

  5. ernest says:

    “My comment relates to servers, not desktops. Most websites run on servers, you know.”

    Your comment appears, mysteriously, to have disappeared.

    Could you do the likes of me and Gewg_ a favour, and re-instate it?

    From memory, it contained the preposterous statement:

    “Since most people use that other OS and one or more popular browsers, the web developers use the same. Often web developers are not system administrators so they don’t need/want many of the advantages of GNU/Linux…”

    Which I’m sure you are more than up to defending.

  6. ernest wrote about users of IT, “they happily go along with building a million dollar operation on the back of a ten-thousand dollar “Microsoft tax.””

    Like Extremadura, Spain, which switched all of its PCs to GNU/Linux? or Munich? or the French national police force, or the governments of Russia, India, China and Brazil?

    I built a system for a single school using GNU/Linux and $40K was saved on M$’s licences. Governments can save $billions using GNU/Linux and it’s not just about licences for the desktops: anti-malware, licences for applications, being pressured to run that other OS on the server to hold the hands of the client machines… It all adds up. On servers, several sources report saving 2/3 the cost of operation and about 1/2 for desktops. In my own experience the savings were greater than that and we had much better performance, really increasing performance/$. The only reason many stay with that other OS is that they have dug a deep hole of lock-in with M$’s help and are unwilling to pay the cost of climbing out. POWs are like that sometimes. They are afraid when liberated.

  7. My comment relates to servers, not desktops. Most websites run on servers, you know. Most client machines are desktops running that other OS. Because M$ built so many deviations from standards into IE, web developers waste a lot of time putting in things like “If IE6 do this instead of that”. About 50% of clients use that other OS and its browser. There’s no good way to test such sites without using IE. W3Schools also has lots on .asp and other M$-specific stuff.

  8. aardvark says:

    So, in other words, you are standing by your comment:

    “Since most people use that other OS and one or more popular browsers, the web developers use the same. Often web developers are not system administrators so they don’t need/want many of the advantages of GNU/Linux.”

    I think most people can live with that.

    I have no doubt whatsoever that most web developers could live with it.

    Not sure where that leaves the superiority of Linux, however.

  9. oiaohm says:

    ch 1991 StarWriter breaks up into stand alone programs.

    1992 3 of the split way programs merge into StarOffice

    1995 all of the 1991 split away programs merged back in.

    1991 is when starwriter becomes a word processor alone.

    That is the problem ch. 1985-1990 Starwriter is a mix of applications in one a suite. 1991 on is just a Word processor.

    So 1995 merge back in is correct. Its a timeline from hell. Starwriter before 1990 was a office suite between 1991-1994 there is a gap. Before Staroffice appears as a full office-suite. Staroffice 1995 basically the Starwriter pre 1991 glued back into on piece.

    You missed the 1991 split. The conversion from Starwriter to StarOffice name was done very messy. Happening over 4 years make it very simple to get the wrong idea by reading the 1995 statement without the 1991 context.

    Of course you were not thinking that the modules take into Star Office in 1995 directly came from the 1991 split up of Starwriter. In fact that is where they came from.

    Yes its one really F up way to change name of product. It gave the impression of being new in market when it was a old product.

    Ch you reason is false for MS rise in fact. The rise is linked to RTF. MS got all other makers to agree to make RTF they got governments to request documents in RTF. Then extended RTF so word would be the only one that would be able to read it properly. This transferred on to .doc.

    This is why MS reacted so bad to the suggestion of open formats.

  10. ernest says:

    “Repeatedly, I am told that it costs less to be locked-in to M$’s way of doing things in IT.”

    Hardly, Robert. You and I are both but fly-specks on the world of commerce.

    Large companies such as Microsoft, Google, Apple, IBM and even (if we wish to descend to mere billion dollar minnows) Red Hat don’t give a toss.

    And, in fact, their customers don’t give a toss, either. Governments, FT500 companies, school districts, even charities: they don’t give a toss about this “lock-in” drivel.

    Mostly, they happily go along with building a million dollar operation on the back of a ten-thousand dollar “Microsoft tax.”

    It’s all about cost-benefit, really.

    Feel free to insist that everybody on that list is making the wrong decision, but, you know what?

    They’ll keep doing so. Life will go on, no matter what either you or I say.

  11. I am in a different timezone. I also sleep and have a bunch of projects unrelated the blog underway. The spam filter is almost perfect in keeping real spam out of my blog but it randomly puts a bunch of ordinary comments into the SPAM cue. I don’t know why. I hope to fix that soon.

  12. Thanks for that. FLOSS being global, I find Google Translate limiting. For some reason Google Translate seems to have a very hard time with German. The output sometimes is not readable in English. Limux even messes up Google Reader. After translating the German newsfeed, Reader gets into a state where it translates English to English and messes that up. I have to reload Reader…

    The point still remains that FLOSS is not always following/copying the lead of non-FREE software. Everyone shares from everyone their ideas about software. FREE software does get to see the user-interface of non-FREE software but of course does not copy code. StarOffice and M$’s stuff were developed in a similar time-frame when people were figuring out how PCs could be glorified typewriters and calculators. For many purposes one is not better than the other but when it comes to price/performance, and freedom, FLOSS is a clear winner. This string of comments stemmed from the assertion that FLOSS copied the non-FREE software and/or was not “original” whatever that means. It’s like the current argument about APIs and their implementation in Oracle v Google. APIs are not copyright-protected works because they are just ideas about how software should work, not an implementation/a work.

    There is a huge base of users of M$’s office suite who are more familiar with the user-interface of LibreOffice than they are with “the ribbon”. That’s not because FLOSS copied non-FREE software but because menus/icons are reasonable and workable. It’s a case of non-FREE software losing its way trying to distinguish itself from itself by unnecessary change in order to sell more licences. I think Ubuntu has similarly lost its way but Ubuntu does not represent hundreds of millions of users, yet. With FLOSS, one can easily dodge whatever errors Canonical may make and still use one’s computers… With non-FREE software it’s M$’s way or the highway.

  13. ch says:

    Mr Pogson, could you please rescue my last post ? Thank you.

  14. ch says:

    Full disclosure: I am a German, so I have the unfair advantage of being able to read and understand that German WP entry.

    Star Writer was _not_ an office suite but rather just a word processor. By adding the other stuff, it became an office suite later:
    [quote]
    es wurde bis 1995 zu einem vollständigen Office-Paket mit allen im Büro relevanten Programmmodulen ausgebaut
    [unquote]
    Translation: Until 1995, it was extended into a full office suite with all relevant modules.

    I didn’t mention Works, because all-in-one programs like that are really a different categorie. MS Works was developed as a cheap home product (for people who couldn’t afford the “big ones” at the time) and mostly sucked, like the other all-in-ones (e.g. the early Lotus Symphony/Jazz and Ashton-Tate Framework), but I liked the “database” part of it: Extremely limited, but for what it did it was extremely simple to use, too.

    MSO was a huge success because it contained Word (a full-fledged program that sold wel at the time because the main competition, WP for Windows, was so bad at the time) and Excel (equally successful because the main competition , 1-2-3 for Windows, was so bad at the time) plus Powerpoint (only so-so compared with the competition, but essentially thrown in for free). That last point (PP for free) was the real reason for the office bundle, of course: It was designed as a Harvard Graphics/Lotus Freelance killer, and it pretty much worked. For this purpose, just throwing the install discs for Word, Excel and PP into one box was sufficient. Yes, 4.x was the first fully-integrated version.

  15. oiaohm says:

    ch I can understand the problem.
    http://de.wikipedia.org/wiki/StarOffice

    StarOffice was not the first name of Star-divisions Office suite.

    StarWriter is the first name of the Star-division Office suite. Later on StarWriter was rename StarOffice and then the name reused.

    But just to make it really confusing 1991 StarWriter breaks up into stand alone programs.

    Only to glue back into one huge program in 1992 for the first release as StarOffice ie 1.0. Inside 1 year it was basically found to be stupid to split StarWriter.

    Just to be really horid StarWriter on windows in 1996 starts counting from 1.0 again

    StarOffice historic is down right confusing that names have been used many times for different things.

    That some versions of Starwriter are direct blood line of StarOffice and some are not. StarWriter for windows basically dies out none of it code base exists in Libreoffice today.

    Yes 4 versions of Starwriter for dos are not accounted for. They happened between 1985 to 1990. Those are particularly important to know when calc, base and draw were created.

    Simple question how many programs is Libreoffice/OpenOffice/StarOffice/Starwriter.

    When you get right down to the nuts and bolts. Writer, Calc, Impress, Base and Draw don’t exist as programs. What appears to be .exe files are basically fancy link files to soffice the real program. This is the way it always has been right back into StarWriter 2.0 for dos in that 1985-1990 time range.

    So from the start StarOffice has been a integrated office suite before it was even called StarOffice. Not a stack of applications bundled and called an office suite.

    1989: MS Office released for Macintosh. This here is not really a Office suite since at this point its just a bundle of applications not designed to work with each other.

    The Microsoft Office for Windows 3.0 is the first to start to show intergration. Note people are still using MS Works at this stage because integration works in MS works where it does not in MS Office yet.

    Yep Microsoft Office 4.0 is the first integrated Office suite under the name Microsoft Office. 1994 that is when it changes from a bundle of applications to something that can truly be called a Office suite. 2 years after StarOffice reform after the starwriter split. Many years after StarWriters first Office Suite form.

    1987 is Microsoft first Office suite by the way. It is not called MS Office but MS works. But MS killed that off. Still StarWriter is still a office suite officially before 1987.

    MS Works is a similar design to StarOffice relations.

    What makes the back track so confusing is all the changes in names and items from MS being called Office when they are not a Office Suite by the define.

    Ch I guess you would have never found the MS Works as a Office suite without a person like me pointing it out. Same problem you have trying to read StarOffice history too many name changes to just trace the name.

    Yes starwriter was first produced in 1984 and formally released in 1985 that goes on to be StarOffice then on to be OpenOffice and finally Libreoffice. There are still code fragments from 1984 in Libreoffice today. Some are fairly important to the design of it.

    The history of one program many names.

  16. ch says:

    > StarOffice which was later released as OpenOffice.org was produced in 1984, before M$ was making an office suite (1990). M$ had a separate word-processor in 1984. In 1982 they had Excel as a separate product. <

    Completely wrong. The actual timeline:
    1982: MS Multiplan released for DOS.
    1983: MS Word released for DOS.
    1984: Star Division founded.
    Apple Macintosh released.
    MS Word released for Macintosh.
    1985: Star Writer (!) released for CP/M.
    MS Excel released for Macintosh.
    1989: MS Office released for Macintosh.

    I couldn't find an exact date for the release of the first Star Office, but it seems to have been closer to 1995.

  17. oiaohm says:

    Ivan Note samsung is just a one. HP Bother Xerox …. All major brands will do it to you sooner or latter.

    Notice Ivan defence it has to be Samsung. In fact its not. Samsung is the most likely of all to work out box. Sorry Ivan I trapped you. Only a particular model Samsung has a issue that is a firmware fix. HP and Xerox have many without firmware fix. You just proved you did not know the topic. And are most likely a paid troll by someone.

    Sharing the scan functionality or just the print function. Sharing the print function I can do without issues by network. Sharing the scanning function by network that is where things turn evil and you run into the issues of needing ftp servers at times to address weakness in firmware of printers that is never going to be replaced. Some physically cannot be replaced because they are rom in the processor chip.

    Ivan
    “Said the person arguing strenuously that we shouldn’t be generous with our code. You should just admit that it’s about controlling what people do, because it certainly is not freedom unless you choose to redefine what is and isn’t free.”

    No you don’t get it. No one has freedom. Everyone has restrictions. Like I cannot just come and shot you because I feel like it.

    Generous is one thing. Being a foot mat to be crushed into the ground is another thing.

    Ivan the define of freedom you point to is not reality of the world anywhere. Because as soon as you try giving people in an area the true define of freedom greed will take over and destroy it.

    Wine project was very idealistic like you when it was MIT license. So was I. Watching the life crushed out of the project by Greed taught me that what you are point to can only exist temporarily until its destroyed.

    Please note MIT license code goes back to the start of Unix. Most of the projects that started the Unix world no longer exist due to the crush force even that there source code still exists in modern-day Unix systems.

    Greed destruction is not new. GPL was design in a time when the destruction was a full force. Today we seeing BSD and so on project grow in number. But If you look at history this happens over and over again. Then the bubble busts and the parties start fighting with each other and the licenses like LGPL GPL are required to come back in to prevent complete failure.

    Then after a while people complain that GPL and LGPL are too strict and the cycle starts again. Its a never ending loop.

    The GPL and LGPL stuff lives through the loops without disruptions.

    IVAN explain to me how if freedom is so great could codeweavers could have protected themselves from destruction by transgaming and remain supporting wine as MIT license or some other open license you agree with and not go LGPL.

    Freedom you are talking about gives Transgaming the freedom to execute codeweavers and not even look back. Yes the wine project gets requests to change from LGPL to BSD and others all the time. Mostly because people don’t consider for one second the wine project has a hostile enemy.

    Transgaming will take the wine source and not give back if they are given a chance. Even go on a marketing campaign to take all of code-weavers clients so removing funding from wine.

    We live in a world that needs money.

    If you can design something that can protect codeweavers in this real world example of hostility and still have a higher freedom license codeweavers is interested. Codeweavers would love to provide max freedom as far as they been able to find LGPL is the most free license possible with the hostile threat.

    IVAN you are a dreammer. The idea has to be workable. This is also a Troll path. Copyleft is not enough freedom is a troll arguement or a under researched idiot. History tells you why copyleft is critical. We don’t have a better system yet.

    Here the screams form those who don’t like copyleft we don’t want to release source. They also don’t want to take part developing shared product either.

  18. Ivan says:

    You belief is not true of you have tested broad enough.

    Never had a problem with XP, Vista, or 7 sharing an all-in-one.

    samsung

    There is your problem.

    The power of greed is far to strong.

    Said the person arguing strenuously that we shouldn’t be generous with our code. You should just admit that it’s about controlling what people do, because it certainly is not freedom unless you choose to redefine what is and isn’t free.

  19. oiaohm says:

    IVAN
    “Permissive licenses prevent unscrupulous, self-funding lawsuits that drive future users away.

    If your entire goal is to let people use your software unconditionally, then let them use your software unconditionally. Don’t sue them into submission.”
    Pick Users or Developers.

    Permissive Licenses you might have Users but no Developers so you project becomes obsolete trash.

    Let them use your software unconditionally is the problem. They take extend and never give back so a competitor from them is forced to take extend and never give back to complete with the one doing the wrong thing. Soon you have all people who would be submitting code to the project doing mine mine mine and the project dies.

    Developers to submit code to a project need some insurance that they are not going to end up in a mine mine mine problem.

    BSD failure before Linux existed and GNU success directly links to the mine mine mine problem that most people pushing Permissive licenses wants to avoid.

    “Possible unscrupulous, self-funding lawsuits” vs “Unscrupulous companies with the goal to dominate everything including destroy other companies working on your project by any means they can” That is what you are choosing between.

    Possible unscrupulous, self-funding lawsuits are not a problem if you obey the license.

    IVAN how are you going protect a permissive license project from having its supported competed in market place out of existence. Leaving behind only the companies that are Unscrupulous companies.

    Lose of a few users here and there because you have not chosen a highly permissive license is offset by the fact you can rest assured that the companies submitting code to your project are not going to find themselves in a location code-weavers did without legal way of hitting back.

    The lawsuits you see from FOSS are mostly about defending developers who are submitting code to the project from being undermined by other companies taking the code and extending and not giving back.

    Users have very little long term bearing if a project will last long term. Developers are more important than Users if you want project to last.

    “software unconditionally” Remember wine was MIT License is an unconditionally license. Codeweavers was facing where they would be force not to provide code to it any more so they could compete against an company also doing the same thing. Thinking Codeweavers hosts the wine website and pays the lead developers full time. They close shop on Wine there is no more Wine project. It was change to LGPL or shut wine down forever when face with an unscrupulous companies.

    This is the front line reality. Many projects using LGPL due to the competitors in the market against them if they changed to BSD inside a year or two they would be a dead project. The limitation of those licenses is preventing death.

    “software unconditionally” its not possible in any case that a many companies are making profit from that software. The power of greed is far to strong.

  20. oiaohm says:

    Ivan
    “That’s nice but what does that have to do with helpful error messages?”
    The changed design allows the error generation to see the exact code that caused the detectable error.

    The old gcc has a forward going process in the complier deleting the history of where its been so templates and other things basically make it impossible to gcc to geneate clear messages why. Yes the new design for gcc is to address that. Fitting it is not a simple process.

    Ivan
    “In your words, drop this before you get hurt, I’ve used multiple wireless all-in-ones from multiple manufacturers with no anonymous ftp required.

    Never had a problem with sharing the output.”

    Note I don’t support the anonymous bit. There are many wireless all-in-ones out there that break when linking to XP or 2000 or to windows 7 pro on a secured network. I have a samsung fairly new and wireless. Hates XP for smb(windows filesharing) Has no issue with Vista or 7. Works perfect with FTP and FTPS. Works perfectly with XP if you directly connect the printer by USB to the XP machine. By wireless or network you better know how to setup FTP or FTPS server. It does support finding those by dynamic dns system designed by apple.

    You belief is not true of you have tested broad enough. At times users will have no option bar use the FTP option. Also with windows updates SMB protocol can change to be incompatible with the multi functions contained version that might not be updatable. FTP and FTPS is fairly much written in stone.

    Drop this idea before you get hurt. You will find the way you are hooking those multi function up is at risk of failure. There are limited dependable ways.

  21. oldman wrote, “We had plenty of choices in applications Pog. That is IMHO all that counted.”

    Nope. A PC is a PC. It is a general-purpose computer. Restricting it to run only that other OS is throwing away a lot of value and good price/performance. We read that many large organizations can do all the IT they need using a mix of ~80% GNU/Linux and 20% that other OS. Some go 100% GNU/Linux. Those organizations save tons of money in lowered licensing fees and lowered maintenance costs. Even M$ is advising governments to use a mix because it has lost the battle of monopoly. If your organization is not using a mix it is clearly paying too much for IT and/or getting too little performance/$.

    Where I last worked, when we switched to GNU/Linux from XP, malware ceased to be a problem and we ran without scanners except on the web connection. Downtime disappeared except for rare hardware failures. Performance improved. It cost less to switch to GNU/Linux than maintaining that other OS and then after switching maintenance was far less effort.

  22. oldman says:

    “If M$ can bully an OEM, what chance do retailers and consumers have of real choice?”

    We had plenty of choices in applications Pog. That is IMHO all that counted.

  23. Clarence Moon says:

    If M$ can bully an OEM, what chance do retailers and consumers have of real choice?

    For a fan of the USA vs DOJ archives, you offer a very biased and misleading presentation of the facts of the matter, Mr. Pogson. Compaq had acquiesced to the Microsoft offer to receive a discount on Windows as a Microsoft partner based on certain actions, one of which was exclusive offer of IE as a browser in the OEM’s package. Compaq took the deal and received tens of millions of dollars in overall discount in return.

    Some underling in the organization failed to heed those terms and tried to put Netscape on some models. Microsoft pointed out to his bosses that such a thing would result in a sizable charge to Compaq. Compaq quickly reversed its error and all was well.

  24. kozmcrae says:

    Clarence wrote:

    “FOSS software is a copycat effort on something made popular as a commercial work.”

    Not true. Software was, for the most part, freely shared in the early years of computing. It wasn’t until later on that companies realized that, “Hey, there’s money to be made here if we can lock it down.”.

    You fail again Clarence in your attempt to revise history. Although I doubt that you really even know that much about history.

  25. Clarence Moon wrote, denying history, “People have always used the best technology available to them to do the job at hand.”

    Nope. M$ did its best to force people to use M$’s technology whether it was good, bad or indifferent:
    “In June of 1996, Compaq removed the Internet Explorer and Microsoft Network icons and a directory of Internet service providers. The company quickly reversed itself after Microsoft sent a letter saying it would terminate Compaq’s Windows 95 license if the items were not immediately restored.”

    see CNET – NEC, Gateway back Navigator

    If M$ can bully an OEM, what chance do retailers and consumers have of real choice?

    Clarence Moon wrote, “Linux and FOSS in general has, with very few exceptions, been a day late and a dollar short in coming to the market. For the most part, FOSS software is a copycat effort on something made popular as a commercial work.”

    Nope. People create FLOSS or non-FREE software or both. They can do it any time they want. Often FLOSS has had stuff before that other OS and “partners”:

    • Apache (February 1995):IIS (February 1996)
    • StarOffice which was later released as OpenOffice.org was produced in 1984, before M$ was making an office suite (1990). M$ had a separate word-processor in 1984. In 1982 they had Excel as a separate product.
    • GIMP was released in 1996. A lot of digital imaging technology was developed at JPL/NASA in the 1960s and is public domain so not a lot of innovation was left by the time the PC came to be. PhotoShop released in 1998 certainly did not do much that was new and wonderful.
    • M$’s media player was released in 1991. OpenSoundSystem for *NIX was released in 1992 and was completely independent of M$.
  26. Clarence Moon says:

    I think that’s all nonsense. Doing anything in IT with M$ underneath puts toll-booths on everything and prevents using the best technology to do the job.

    Au contraire, Mr. Pogson. Rather, what you suggest is nonsense. People have always used the best technology available to them to do the job at hand. Perhaps the term “best” merely meant “most accessible” or “already on my computer”. Or perhaps “what they had at the store” or “what my neighbor used” or even “what the boss gave me”. In any case, the user had the ability to choose something that got the job done in what seemed, to the user, to be an adequate manner.

    Linux and FOSS in general has, with very few exceptions, been a day late and a dollar short in coming to the market. For the most part, FOSS software is a copycat effort on something made popular as a commercial work. FOSS efforts range from excellent to poor productions in terms of usability. Once someone has an acceptable solution, there is no advantage in having another and what one already has is just as free as what someone wants to give away. Free-er, in fact, since there is no additional learning curve or installation effort needed to simply continue with what one has.

    When it comes to choosing “the best technology to do the job”, the issues are not just price and some obscure banter about bits and bytes. FOSS has to lead the market and, so far, it has never done that.

  27. Ivan says:

    LLVM/Clang is in fact based off a GCC long term plan document.

    That’s nice but what does that have to do with helpful error messages?

    lots of fear mongering about permissive licenses

    Permissive licenses prevent unscrupulous, self-funding lawsuits that drive future users away.

    If your entire goal is to let people use your software unconditionally, then let them use your software unconditionally. Don’t sue them into submission.

    So may work random-ally.

    In your words, drop this before you get hurt, I’ve used multiple wireless all-in-ones from multiple manufacturers with no anonymous ftp required.

    Never had a problem with sharing the output.

  28. oiaohm says:

    Ivan
    “But X-Server keeps on crashing.”
    Only true on miss matched hardware using old driver designs. DRI2 drivers rarely crash. Ok not as functional but dependable.

    Ivan
    “Yes it is. If you must use a Unix-a-like the permissive licensed tools, which meet your cult leader’s redefinition of freedom, provide a better experience. ref: error messages from LLVM/Clang v. error messages from GCC.”
    Lets stop and look very carefully. Why your statement is a sign of a fool. LLVM/Clang is in fact based off a GCC long term plan document.

    Why is LLVM head of GCC at moment GCC is having to convert a legacy code base LLVM started clean. That is all the difference is.

    Where there BSD based compliers before LLVM yes there were.

    Where are all those bsd compliers dead or dieing. Why did they die. Closed source complier took there source base and extended and never gave back. Yes the Unix compliers that got displaced by Gcc over time all had a common blood relation in MIT license.

    Too permissive of license is the path back to the Unix mess with forks off all kinds of directions and no true compatibility.

    I saw where the permissive license leads once there is money to be made. Wine vs TransGaming. Remember wine started as a MIT licensed project.

    TransGaming decides it will take the wine source code and give nothing back. Code-weavers and other major funding firms of wine find themselves in a desperately bad location. Any new feature Code-weavers release on wine TransGaming will pick up instantly and TransGaming is giving nothing back worse TransGaming states that if wine dies it will be good for their business.

    Wine changes LGPL to protect Code-weavers and other companies funding from TransGaming aggressively taking code. Yes this was a case of either shut the wine project down or change license due to aggressive assault by TransGaming. Why because code-weavers to compete with Transgaming would have no option bar to keep there future alterations closed source so death of wine. Years latter TransGaming falls behind and is forced to convert more and more of there code base. Reason they did not have the resources to maintain it. Only had the resources to destroy.

    This is not the first time this has happened or the last. Utah GLX was another case of aggressive assault this time Nvidia was the partly doing the aggressive assault so Utah GLX collapsed.

    Core to many common programs in Unix systems also collapsed in quality due to aggressive assault. So almost none of the Unix vendors were releasing features and code back to the BSD projects where the source of their tools was starting.

    LLVM would be fully open to a full out aggressive assault. Biggest developer on LLVM is Apple if Apple decide to turn against the rest of the LLVM community the battle could turn very savage.

    Permissive licensed tools is a road to ruin. Its just a matter of time. Really I would have just sat back myself let the busybox clone go forwards and slowly implode on itself like the Unix environment did over the years. Today when you look at solaris and aix and other unix large sections of there user-space is GPL and LGPL or there own licenses with equal restrictions against hogging the source.

    Permissive license falls into the camp of be careful what you wish for you just might get it.

    Result of getting a highly Permissive licenses is accepting that if developer groups in the project are competitors to each other they might get in a blood feud resulting in the absolute death of the project unless the project stops having a highly Permissive license.

    LGPL, MPL and CDDL seam to be the most Permissive License you can go before it becomes too far open and you risk competition based destruction to the project.

    People asking for BSD, MIT ….. Mostly have not checked history of what happens to project using those licenses once competition over market share kicks in. Instead just thinking it will be great to bind into my project and not have to give back. Of course forgetting if that leads to the event of no one working in the project it dies.

    Ivan
    “They don’t want to have to dick around with anonymous FTP for sharing the output of a scanner on the network.”

    Drop this before you get hurt. Network scanners normally only give you 2 options ftp and smb. Problem is SMB may never have been tested with windows at all. So may work random-ally.

    So you want dependable network scanning you have to setup a ftp server to receive it hopefully the printer supports ftps what is encrypted.

    So this is not a case if user having a choice. They have no option bar to setup ftp of some form. Setting up anonymous ftp in this day and age I do have a bone to pick over.

    There is no common standard for sharing scanner over network other than ftp dumping the results. This is also due to the fact you don’t want a scanner puppeted remotely from anywhere on earth. USB you can computer control scan. Network no you cannot. There is nothing that makes setting up network scanning simple. Yes as users I can understand being upset over this. Reality is a true ass a times.

  29. Ivan says:

    malware doesn’t happen

    Yet you say this about your platform quite often, everyone knows the line is false.

    M$’s stuff is just better and easier to maintain

    You’re using anonymous FTP for network scanning in 2012, you should be complaining about that and wanting an improvement of the situation, but here you are, deflecting.

    more/better applications are available for that other OS

    Yep.

    GNU/Linux is bad, the GPL is bad,

    Yes it is. If you must use a Unix-a-like the permissive licensed tools, which meet your cult leader’s redefinition of freedom, provide a better experience. ref: error messages from LLVM/Clang v. error messages from GCC.

    And let’s not forget the self funding lawsuits of Busybox that were caused by the license and the unscrupulous people of the SFLC. How did those work out? Oh yeah, the person that started them was flamed on LWN for working on a replacement.

    users want M$ so you are better off making them happy

    Users want stuff to work. They don’t want to have to dick around with anonymous FTP for sharing the output of a scanner on the network.

  30. iLia says:

    Doing anything in IT with M$ … prevents using the best technology to do the job.

    Please, name at least one such technology.

    fewer re-re-reboots

    But X-Server keeps on crashing.

    GNU/Linux is bad, the GPL is bad, no one makes anything good for GNU/Linux

    — GNU is bad
    — GNU/Linux is bad
    — Linux (kernel) is good
    — almost no one makes anything good exclusively for GNU/Linux

    users want M$ so you are better off making them happy

    Oh yeah, users want MS stuff and are ready to pay, if they have money, or, if they don’t have money, they are ready to pay for a pirate copy.

    because there’s just less software getting in the way of what you want to do

    Oh yeah, there is so few software in the Linux universe that users very quickly switch back to Windows/Mac OS in order to be able to do what they want or what they need.

    Of cause, if they want to configure linux they can stay with it and be happy.

  31. Phenom says:

    The costs of lock-in are high, indeed. This guy can justify it. Gosh, he was even a paying customer:

    http://ehsanakhgari.org/blog/2012-04-14/how-i-lost-access-my-google-account-today

Leave a Reply

Your email address will not be published. Required fields are marked *

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <strike> <strong>