Lost Sheep

First ASUS unleashes netbooks with GNU/Linux. Then they push that other OS. Now they are selling good netbooks with no OS but Expressgate. What’s with that?

Surely M$ does not approve. Is this another sign that M$ is losing its grip on OEMs? You can find “No OS” products from ASUS all over the world. Finally, an OEM is recognizing that there is a market for computer geeks who are willing to do a bit of work for $50 off the price. The solidarity of partners is now in question. If the partners can make as much money by not sharing with M$, why should they? The result is more choice for consumers. Amen. Is ASUS responding to the market or behaving like a lost sheep needing rescue? Will M$ “rescue” ASUS by giving them still better a deal? Will the market decide what to buy based on price and performance? Stay tuned.

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.

18 Responses to Lost Sheep

  1. Amen. If GNU/Linux can run rings around XP on old hardware imagine what it can do on modern stuff? My project for today is to set up a brand new machine with 4gB RAM and dual 500gB SATA drives in RAID 1 with a dual core 64bit CPU. I will install VirtualBox and play with recording screenshots and video of operation in the virtual machine for training students and staff. I started last night with an erasure of that other OS. I write this while the drives sync.

    I am not an engineer but a scientist by training. Education is my fifth career. I understand how a lot of things work and how to solve all kinds of problems. I have worked with many kinds of hardware and software over 40 years. Ad hominem attacks do not cause me to lose sleep at all. I do take offense at the boring repetition of lies and half-truths from sycophants of M$, hence the tone of my blog. Readership is a growth curve…

  2. oe says:

    Ad hominem attacks usually indicate that little substance of objective merit is in the arguments of a party; and I repeat:

    “I can only presume that you are not an engineer, Robert. Nor do you seem to be much of a software developer. You are just huffing and puffing here with no direct knowledge or understanding.”

    I don’t need to speak of processor caching, hit/miss rate, bus addressing speeds, etc. or even know those internals, when my new netbook ran like a dog with that other OS and yet the same hardware, an ASUS MKH1000, after to borrow a term, after being “paved over” with Easy Peasy (a Ubuntu/Debian derivative) runs pretty darn slick. Matlab2008 runs just fine on it, Octave (a GNU Matlab clone), WxMaxima, and Open Office Calc/Writer/Impress, LyX all run snappy. The subjective user experience has been enough to convince me without needing to go into the intricacies of RISC/CISC, branch prediction, caching. It’s the same story at my work desk, the older generation hardware (CentOS 5) is where 95% of my work is done, because the new issue hardware, running That Other OS (with 4X the RAM and about 2X the CPU processing power) is like a slug stuck on glue with the HDD always thrashing away with malware scanning, updating (especially on Tuesdays) and heaven knows what else.

  3. I am usually unpaid for my tech work. I am a teacher. I need tech to work to do my main job so the cost to the employer for my GNU/Linux knowledge is minimal. My salary here is around $70K so if we divide that by the staff and student population it still comes to less than one licence fee for that other OS. Today, a complete newbie got one of the new (hot) machines. The only help I had to give him was to tell him to type in his user name. He was used to that other OS with icons … At the end of the day he told me it was if the machine was reading his mind. Google Chrome-browser does that, more or less and the machine was fast. We do not need the scan-on-access stuff to slow it down. This machine has two drives but the second one needs a SATA cable to do RAID. Then he will know what speed is. Over the years I have introduced thousands to GNU/Linux. I have been installing GNU/Linux in schools where I taught for a decade and moved to a new school almost every year. Most of the systems I installed are still running.

    WGdisA is a very intrusive thing and here, on XP, it insists on ActiveX being enabled which is a serious security risk. Their whole OS is buggered by the requirements they put on the user, the machine and the network. It’s just not efficient. The machine I installed today is doing nothing but working for the user and the user appreciates it.

  4. amicus_curious says:

    “I converted lots …”

    Sure, Robert. How many was that, anyway? A hundred? More? Less? Divide your annual salary by that many people and see what an unjustifiable cost is part of any sort of Linux conversion. And that is annually, paid over and over again for the “support” needed to maintain Linux. OTOH, people self-administer Windows. It is true that they don’t do such a hot job and fall prey to lots of malware in their innocence, but they feel as though they are in charge of their own fate and that is more important to them than the dweebishness promulgated by the Linux fans.

    Consider that much of the respect people may show you over the Linux issue is simply due to not wanting to disturb or argue with a crazy man.

    “What’s with that?”

    WGA isn’t the same thing as a checksum, Robert. It is a credible effort to protect Microsoft against copyright fraud, which lets them pick and choose when and where they may want to prosecute violators and even serves as a notification to users who might care if they got a genuine copy of MS software after paying for one. Admittedly it is not much of an obstacle to knowledgeable pirates, but such folk are few and far between and there really isn’t any true revenue loss from such activities. The warez crowd would not pay if they could not successfully convert. They would simply do without or even get by with FOSS stuff.

  5. Nonsense! Munich is under-budget. Many others report the amount of training needed is far less than planned. I converted lots of teachers with very little except showing them how to login and fire up the usual apps. Migrating a large organization from XP to “7” is pretty costly, too. I recommend looking at this from the perspective of what it costs not to migrate. Mal-ware, slowing down, re-re-rebooting and licence fees forever is a huge cost that should not be ignored.

    Telephones are computers these days. Some people use them more than they use a PC.

    Have you noticed that you must have ActiveX enabled, use IE and WGdisA to authenticate boxes these days? What’s with that? Way too much complication if all they need are some checksums.

  6. amicus_curious says:

    “How about the entanglement of the browser with the OS and Active X and all that?”

    you don’t make much sense with this, Robert. Certainly the IE Browser is an application that runs on the Windows platforms and uses the Windows APIs. Some 15 or so years ago, some of the elements used by the browser were in DLLs that contained other platform functions, but that was changed a long time ago. Where have you been hiding?

    “The licensees sell the chips in the billions.”

    But they stick them into a telephone, Robert. If/when they start sticking them into a computer, they will doubtless be interested in a version of Windows suited to the architecture they use. Windows has been on RISC processors in the past although those processors ultimately succumbed to the x86. There was a reason for that and not much has changed in the interim.

    “I am grooming a PC …”

    Well, goodie for you, Robert, but don’t let this one on one sort of effort blind you to the immensity of the task of converting a large organization or even many individuals to Linux. Large scale efforts, such as you reference in Munich, have by and large been very costly and can only be seen as commercial failures due to the slow progress and added expenses encountered by these rash folk who believed in the FLOSS story to their chagrin.

  7. How about the entanglement of the browser with the OS and Active X and all that? There is no need to have such a close relationship between browser and OS, which is why I call it improper. This tangle was created to block NetScape, not for any technical imperative.

    ARM sells designs not chips. The licensees sell the chips in the billions. The people who have already migrated to GNU/Linux were not willing to wait. There are many millions who are in the process as is my employer. I am grooming a PC for a teacher who asked to try GNU/Linux for tomorrow. Once they see the possibilities there is no going back to life in a straight-jacket.

  8. amicus_curious says:

    “To argue that “7” or “8” will do well in ARM performance-wise is to suggest that there is no bloat in that other OS. Bloat I see every day: improperly nested dependencies…”

    What improperly nested dependencies do you see in Windows 7, Robert? LOL. You are speaking gibberish butI think you know that.

    “No, M$ is not going to convince ARM to support the weight”

    ARM, just as Intel and AMD, are in business to make money and if there is some accomodation needed to make their products sell at higher volume, you can expect that they will make such changes. People want a better Windows far, far more than they want a complete change, Robert, and they are clearly willing to wait as long as it takes. That has been the history of this market for the past 20 years and your simplistic view of its dynamics is totally wrong.

  9. How large a CPU cache does “7” need to operate well? How large are the caches on ARM CPUs?

    For those that do not know what caches do: see http://en.wikipedia.org/wiki/CPU_cache

    The bloated innermost loops of Vista needed 2 MB caches. “7” is not much better. ARM CPUs often have <512KB caches.

    The CPU caches are on-chip and operate several times faster than main memory. e.g. 20 gB/s compared to a few gB/s.

    The reason ARM CPUs do not have caches nearly the size of x86 CPUs of the current era is that ARM instructions are smaller and you can get more instructions in the cache at once. Even if translating that other OS to ARM is possible, the bloat will not fit. To argue that "7" or "8" will do well in ARM performance-wise is to suggest that there is no bloat in that other OS. Bloat I see every day: improperly nested dependencies, malware, scanners, WGdisA, etc. For that other OS to fit in ARM systems well, the surgery would have to be serious, akin to a rewrite and we know how long it took to re-write Longhorn Vista. How many years was that? Six, Seven? A work still in progress? “7” clogs up Atoms pretty well and they have 512 KB caches. Oops, the latest have 1MB caches. I wonder why… Could it be that Intel is trying to help “7” with its bloat problem?

    CPU caches accelerate operations generally until the innermost loops do not fit, then they actually give an overhead as they thresh being constantly refreshed, just like swapping/virtual memory.

    I enjoy the difference between GNU/Linux and that other OS when it comes to processes. A GNU/Linux system can run many more processes on a machine because of shared memory. That other OS will have multiple copies of code that are unnecessary. GNU/Linux only needs one copy of large programmes like OpenOffice.org writer to satisfy all the users on my terminal servers while that other OS will have multiple copies. That other OS may have 20-50 processes for a single user in X MB of RAM and be swapping like mad whereas GNU/Linux may have more processes but not swap at all.

    No, M$ is not going to convince ARM to support the weight. They could, in principal, license the technology and design ARM chips that would work for that other OS but I doubt M$ would get into that. They are all about software and sales and arm-twisting, not CPUs. M$ boasts about spending $8 billion per year on research but almost all of that is just designing their silly software to corner the market. It would take them only a couple of billion to have their own ARM fab but who would buy their stuff? M$ makes its profits by selling stuff at outrageous prices. They could not make it producing their own chips on narrow margins just to get a piece of the ARM marketplace.

  10. amicus_curious says:

    “ARM is a barrier to that other OS”

    I can only presume that you are not an engineer, Rober. Nor do you seem to be much of a software developer. You are just huffing and puffing here with no direct knowledge or understanding. Stick with what you actually know and quit making such silly statements.

  11. IBM made an exclusive deal with M$. People bought stuff from IBM with PC-DOS. People made software for DOS.

    This timeline shows nothing much until 1987 and M$ squelched it by making sure apps designed for MS-DOS would not work with the competition. It was not until 1993 that IBM split with M$. By then the monopoly was solid. The deal with IBM gave M$ a clear path for at least a decade.

    Intel does not have to die to give GNU/Linux a shot. GNU/Linux likes Intel and vice-versa. Intel as a part of the Wintel monopoly has to fade. That is happening more or less rapidly. ARM is part of that. Moore’s Law is another. The anti-trust ruling is another. The lower the cost of IT, the better GNU/Linux looks because it is the only way to lower prices further. We now see no-OS netbooks. That works for GNU/Linux. VAR or computer geeks can load GNU/Linux on these barebones machines with no competition from that other OS.

    ARM is a barrier to that other OS. The ARM architecture does not have huge caches that that other OS needs and ARM does not do bloat very well. ARM demands an efficient OS like GNU/Linux.

  12. amicus_curious says:

    “M$ prospered because IBM gave them a monopoly on DOS on IBM PCs.”

    That is a silly thing to say, Robert. For one thing, PC DOS was not exclusive on the original PCs and was not even bundled with the PC itself. It had to be ordered seperately. And IBM PCs themselves were not the universal sort of product that they are now. The far and away market leader in personal computers was Apple and Tandy was second. I would imagine that the S100 bus machines using CP/M were collectively third. The IBM PC took over the market as did Microsoft OS in a direct competition with the best of the best of the time. You conveniently forget that.

    “The market never made a choice because there wasn’t one.”

    There has always been a choice, Robert. The market was free then, as they are today, to accept or reject what is offered. You say that the market will accept and even embrace the clumsly and ill-finished Linux platform in lieu of Windows in order to save a few bucks, but that is the exact same market that rejected alternate platforms such as CPM-86 and OS/2 in the past and has continually rejected Linux in the last decade when it was offered on low-end PCs by WalMart and others. They rejected Linux when it was offered on Compaq and HP machines and a variety of Dell initiatives. They rejected Linux on the netbook, too. And with all that history you still count on the market to suddenly turn to Linux? It will never happen.

    You keep trying to rescue your theory with the notion that the market really never had a choice, but please explain how the market is even likely to hear of an alternative, much less choose it today.

    FLOSS fans have been chanting “wait til next year!” for a long time now. The current cry is “Wait until the ARM devices are out!” Do you really think anything is going to change? All you have done is added a second massive object to oppose you. Now, in addition to opposing Microsoft, you have to pray for the demise of Intel. Good Luck! LOL.

    “I added just a touch of salt and a smidgen curry to your words a_c to make them a little more palatable”

    LOL, Richard. A trite and childish effort in keeping with the traditions of the FLOSS advocates!

    I agree that Google is a raging success in the mold of Microsoft, Oracle, Intel, and the other modern tech leaders although I have never understood quite why people spend so much money on these click-throughs that Google sells them. Somehow or another public attention has become a market commodity in its own right.

    I don’t see where the cell telephone business has such a connection to the traditional PC business, though. I am not that involved or even that familiar with the cell phone market. I am a sort of cheapskate just like Robert in that regard. I have a couple of T-Mobile phones that I originally set with the $100 pre-pay that is valid for a year and lets me renew annually for a $25 refresh. That gives me cell service for about $3 per month based on the 20 to 30 minutes a month that I use.

    I see others using their phones in an always-on mode and even surfing the web. What I do not understand with Android is just how Google gets any revenue from it. To use it, you have to be a phone manufacturer AFAICT. What is in that for the common man? Perhaps I should buy an Android phone and pay ATT to get service.

    “I wonder when my 16-core ARM netbook will come out…”

    YOUR 16-core ARM netbook? Presumably you are bragging that you will be the first to purchase such a thing. IF it comes out, of course, and that is a very big IF. Do you suppose that, IF ARM machines become a big hit, that there would not be a version of Windows successfully marketed to the world to use on it? Remember the netbook with Atom processors? Why would you think that history would not repeat?

  13. If Moore’s Law applies to ARM, it should be in about 18 months… In this, The Year of ARM we should see dual-core processors for ARM and, I expect that some machines will have two processors, one for stuff and the other for multimedia/display etc. That’s four cores this year, 2010. In 18 months, the resolution of dies should permit 8 cores and there is a possibility for 16 as some people plan a two-step jump for resolution. There could be a move to servery for ARM this year or next. Then multiple cores will be compulsory.

    Competition is good for consumers of hardware. Some people foolishly argue that competition in software and media are somehow not good for everyone but I don’t see it. Even those who love M$ should love FLOSS for providing competition in price/performance.

  14. Dann says:

    The only benefit to purchasing a computer with Windows on it is having all that crapware loaded on it; Companies pay for its installation which lowers the overall price. I wouldn’t mind having those companies load crapware onto GNU/Linux-preloaded machines, as I would wipe it beforehand anyway and get the savings.
    However, machines with no OS is a step in the right direction for consumers.

    I wonder when my 16-core ARM netbook will come out…

  15. Richard Chapman says:

    “…the ineffective whining of the FLOSS folk will never manage to affect mainstream commerce. Never.” It was not “whining” that buried Windows Mobil in Android’s dust, it was:

    -by providing products selected by the market over alternatives.

    -building one of the strongest brands in history and have become almost synonymous with search engines themselves.

    -The term “Google” was invented to define the essence of search itself.

    -careful attention to the market and masterful product management in over a decade of market evolution.

    These are the facts of the matter a_c. FLOSS advocates don’t even bother to market and they are still beating Microsoft badly in mobile (Apple too in growth). In such a huge market as search, with billions of ad clicks per day, the ineffective boasting, lying, bribing, bullying and threatening of Microsoft and their partners will never manage to affect mainstream search. Never.

    I added just a touch of salt and a smidgen curry to your words a_c to make them a little more palatable.

  16. M$ prospered because IBM gave them a monopoly on DOS on IBM PCs. Business bought IBM PCs. ISV provided apps for DOS. When the GUI came along, the apps were migrated to that other OS. The market never made a choice because there wasn’t one. UNIX was $1000 per seat in those days.

    GNU/Linux has affected the market considerably. See Netcraft for effects on the web. On the desktop we have 64bit Flash for GNU/Linux. Why is that if there is no market for desktop GNU/Linux? Repeat the blather all you want. GNU/Linux is making steady strides and that other OS is hanging on by its fingernails. Netbooks and smart-things are just the beginning. People all over the planet see GNU/Linux doing well there. There is nothing to prevent it spreading to the desktop. Business has given GNU/Linux a close look and many are going that way especially with thin clients. The consumer space is holding for M$ for now but the price/performance issue is well established and suppliers of GNU/Linux systems are doing well. I just learned that ASUS is shipping eeePCs without an OS all over the world. Is that not the beginning of the end of the monopoly? ASUS was losing share by shipping only that other OS. If ASUS gains by shipping no OS, others will do the same and then they will get into GNU/Linux again. Where is the upside for M$?

  17. amicus_curious says:

    “Will the market decide what to buy based on price and performance?”

    Well, Robert, that has always been the case. Microsoft has prospered immensely from their providing products selected by the market over alternatives. They have built one of the strongest brands in history ahd have become almost synonymous with personal computers themselves. The term “Wintel” was invented to define the essence of the modern market.

    I find it most curious that you would subscribe Microsoft’s success to having monopoly power and at the same time belittle their chances of future success in competition! You seem to be totally ignorant of the processes involved and believe that the claimed monopoly is some sort of dirty pool trick played by Gates and his fellows on the world and not the result of careful attention to the market and masterful product management over a quarter century of market evolution.

    The facts of the matter are that Microsoft knows how to market and the FLOSS advocates do not even bother to market. In such a huge market as personal computers, with hundreds of millions of unit sales per year, the ineffective whining of the FLOSS folk will never manage to affect mainstream commerce. Never.

  18. Richard Chapman says:

    It must be rather unpleasant for one of a company’s suppliers to dictate what it can and can’t do with its product. Microsoft can and has caused a lot of unnecessary pain in the industry. The whole industry, one company at a time. One Lion can disrupt and scatter 10,000 Wildebeests. It would only take a dozen or so to make a stand and send that Lion packing. A loose consortium of a dozen or so companies would have Steve Ballmer blithering, “Yes sir, no sir… I understand completely sir. Is this going to be on Youtube?”.

Leave a Reply

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