Decisions About Small Cheap Computers

So, M$ is competing on price for small cheap legacy computers these days. That’s OK, but at least you, the consumer, now have a choice you didn’t have a few years ago.“your account information is all stored in Google’s cloud, so you can immediately personalize any Chromebook to your profile simply by logging into your Google account. There’s also a guest account, so someone else can log in, use the Chromebook, and when they log out all that profile information simply vanishes.
This makes the Chromebook a maintenance dream. There are no updates. There are no antivirus programs. There is no maintenance whatsoever except power up and powering down once in a while.”
David Gewirtz has a pretty good piece on the choices. Basically, he suggests, it comes down to what you want to do with your machine. If you absolutely must run applications that need that other OS, pity you, poor slave, then so be it but for everything else, there’s a ChromeBook and one option he omits, installing GNU/Linux on that ChromeBook. You can do that. ChromeOS is basically a browser running on GNU/Linux so it’s already there underneath and if you can write to the storage device or get root access some way, the machine is yours. You can also do that with the freebie small cheap computers bearing M$’s stuff although there’s no need, for the hardware isn’t terribly different. Both Intel and ARMed versions are available.

I would stick with the ChromeBooks lest M$ get any richer/more powerful/abusive. There’s a rumour that $149 ChromeBooks will soon be here. That might still be the deciding factor.

See How to decide between a $199 Windows notebook and a Chromebook.

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 and tagged , , , , , , . Bookmark the permalink.

40 Responses to Decisions About Small Cheap Computers

  1. oiaohm says:

    But I’ve never once heard of anybody with a Windows system in the last fifteen years or more needing to drop down to the CLI and do the equivalent of a modprobe.
    Basically this is putting head in sand DrLoser. The reason why no one has in the last 15 years is it has not been a possibility under Windows.

    http://www.thewindowsclub.com/usb-devices-not-working-windows
    This would be made a lot less painful if you had support for rmmod and modprobe. Find the defective driver unloaded change it settings and reload it is how it works under Linux. These features don’t have to be implemented as CLI. To be correct you have support for a kind of modprobe called hardware detection but its extreamly rough. The unload driver option is missing from windows. Uninstall what is remove the driver completely you have. This is a case that Windows is missing a option. Linux people notice it missing. Windows users are so use to it missing they don’t notice it.

    DrLoser even on Linux to adjust module settings CLI not exactly required. Its everything as a file is a true pain in ass. /sys/module/ Yes that directory contains directories of all loaded modules and their current settings.
    http://sourceforge.net/projects/ndisinstaller/
    Some people have made graphical applets for particular drivers. But these have lacked popularity as it works out a lot easier to give users instructions in command line form.

    This is a case the functionality on Linux is great but the user interface suxs. The interface under windows is ok but is functionality is limited.

  2. oiaohm says:

    DrLoser
    In general, Windows devices (including USB) have their own configuration applet. Which is, naturally, backed by MSI and the Registry. Which is, naturally, an ACID transaction.
    DrLoser read what you said. Changing configuration on fly under windows requires own configuration applet. Linux modprobe does it and linux has a stack of generic interfaces for doing this like /proc. Linux you normally don’t have per driver tools. More generic you driver control is the more audit-able it is.

    In fact bogus claiming that devices that allow all configurations to be changed on fly are in fact using registry. Most Windows Driver Control Applets uses direct messaging to driver to change setting not registry. Writing to registry is so setting comes up next boot. This difference is why a perosn plugs a device in under windows messes with Applet unplugs it and replugs it in and the setting changes disappear then they reboot and the setting turn up again. Its a nice strange Windows behavior that traces to exactly how the driver settings are operating.

    Yes registry has ACID no this does not apply to drivers under Windows completely. Reason is the profile system of Windows drivers. You know safemode loads a different driver profile to default running. There is particular registry mapping that happens when a NT OS boots. Drivers look in the same locations for there settings all the time no matter the boot profile. Guess what happens to be read only and will only be replaced when the system is rebooted. This section of registry also defines if driver signing is on or off so you kinda don’t want this to be played with.

    Audio setting ok. You are having audio cut out at random intervals. You need to adjust power management options on the audio driver. You must reboot as all audio driver power management stuff is define in the profile that you cannot modify while the system is running. Yes you can change your power management profile but you cannot change if a driver will ignore power management options or act specially to power management instructions without a reboot.

    Yes some audio setting under windows are outside of the boot profile. Not all. Yes direct messaging to drivers that Windows driver makers have resorted to is horible. Thousands of different interfaces sending messaged to Windows kernel mode in undefined ways. This behavior does cause security flaws there have been many windows rootkits that have depend on force loading a driver then exploiting its applet control interface to run arbitrary code in Windows kernel space.

  3. DrLoser says:

    DrLoser exactly how do you change module parameters under Windows. That is right edit registry and reboot.

    Unmitigated crap, oiaohm.

    In general, Windows devices (including USB) have their own configuration applet. Which is, naturally, backed by MSI and the Registry. Which is, naturally, an ACID transaction.

    There are certain, fairly dire, circumstances where you need to go into the Control Panel and fiddle around with settings. Audio springs to mind. I have no idea why both the Windows desktop and the Linux desktop seem so determined to cause users pain when it comes to something so simple, but there it is.

    But I’ve never once heard of anybody with a Windows system in the last fifteen years or more needing to drop down to the CLI and do the equivalent of a modprobe.

    That’s such an unspeakably primitive idea that I can’t see why anybody would bother to defend it.

  4. oiaohm says:

    1) No. If there’s any equivalent on a M$ system, it would be an “unplug-install-replug.” But of course you have no cite for this whatsoever, do you, oiaohm?
    http://www.thewindowsclub.com/usb-devices-not-working-windows
    DrLoser you did not read this this contains the process for changing 1 driver option under Windows. Unplug and Replug will not change driver settings. Restart driver maybe change settings no. Modify registry then reboot to have bootloader reload hive with the settings for drivers.

    modprobe [-v] [-V] [-C config-file] [-n] [-i] [-q] [-b] [modulename] [module parameters…]
    This is the modprobe command line. Note the module parameters section. DrLoser exactly how do you change module parameters under Windows. That is right edit registry and reboot.

    Even when I include a cite DrLoser claims I have not. Where is your cite that unplugging and repluging device works for changing module settings. There is no such thing in most current versions of Windows. Its not how current Windows is designed.

    If the bulk of the driver is in user space (as in, post Vista), you might not even need to go that far.
    8.1 kinda ruins this idea. Basically you are not up to date. Microsoft for a security thing is a pain in the ass. A userspace driver only sees the registry as it was loaded by the bootloader under Windows 8.1. There was a few issues with users tweaking driver settings and causing crashes. Microsoft solution make reboot required to change driver settings.

    Windows 8.1 has a change. Vista to 8.0 was nicer due to the Userspace driver stuff.

    2) Here’s the thing about the Windows Registry, when it comes to drivers and such. It is a very good way of storing “state.” And it is backed, for purposes of installation/modification/uninstalling, by an ACID transactional system that is amenable to a declarative workflow.
    Kind and Kinda not true. When it comes to devices that you open up regedit and see that device has X value applied but the driver is using Y value because that is what the bootloader loaded hello confusion.

    Advantage of the Linux method.
    1) new drivers can be loaded at any time will full settings in development systems.
    2) massive what is loaded audit options. This includes methods to find attacks that have managed to directly alter the kernel memory space.
    Disadvantage
    1)No formal signing of drivers.
    Ok a few possible risks but nothing that bad.
    Windows method.
    Advantages
    1) driver signing.
    Disadvantages.
    1) reboot required to alter/install drivers fully.
    2) drivers loading without full settings can crash the OS and this happens due to the reboot requirement to update driver settings the driver sees.
    3) driver signing does nothing against cases of direct kernel space memory alteration.
    4) a Kernel mode rootkit driver does not need settings.

    Of course the arguement is the Linux method allows Kernel rootkits is kinda false. Windows rootkit makers have been able to get Windows signing keys without too much trouble. Its not like a driver cannot have all it setting embedded so undoing the requirement to reboot.

    Difference between windows and Linux over kernel rootkits is the tools for Linux to detect them are free for everyone. Windows you must pay to have tools to detect them as a company.

  5. DrLoser says:

    Are you really a “Microsoft VAR” salesman, oiaohm?

    If so, I worry about your customers in northern NSW. I mean, you appear to have no knowledge whatsoever about M$ products at all, don’t you?

    Some of them are good. Some of them are bad. Some of them are possibly disastrous.

    I’d assume that the very least you could do as a “Microsoft VAR” would be to triage the darned things.

    Oops, another inadvertent cite. I’m a bad boy. Tell Mommy to spank me, oiaohm.

  6. DrLoser says:

    One small gibber out of a wall of gibberish:

    If you require todo equal to a modprobe tweak of driver settings under windows it reboot computer. Driver command line equal options under windows are in registry options.

    1) No. If there’s any equivalent on a M$ system, it would be an “unplug-install-replug.” But of course you have no cite for this whatsoever, do you, oiaohm?

    If the bulk of the driver is in user space (as in, post Vista), you might not even need to go that far.

    2) Here’s the thing about the Windows Registry, when it comes to drivers and such. It is a very good way of storing “state.” And it is backed, for purposes of installation/modification/uninstalling, by an ACID transactional system that is amenable to a declarative workflow.

    In other words, on an M$ system, you don’t need to reboot unless you actually need to reboot. In contrast, on a Linux system pre-systemd — you might or might not. Basically, you didn’t have a clue either way.

    Hello, corrupted state! Oh, did I mention systemd? Here’s the relevant quote on systemd, from the unimpeachable source of Boycott Systemd:

    10. systemd’s complicated nature makes it harder to extend and step outside its boundaries. While you can more or less trivially start shell scripts from unit files, it’s more difficult to write behavior that goes outside the box, what with all the feature bloat.

    Strangely enough, that is actually a benefit of using systemd.

    Just for once, it forces amateur wannabes to write the boot process in a deterministic way. Just like the way that Windows NT has done it since 2000 or earlier.

  7. oiaohm says:

    DrLoser funny the comment about loadable modules not being a stable ABI have in fact blocked a lot of kernel mode Rootkits.

    QA whatsoever for an LKM
    Not true open source mainline Linux Kernel Modules have to pass QA.
    https://sparse.wiki.kernel.org/index.php/Main_Page

    If the LKM is open source and not mainline you can run sparse and other things yourself against it.

    Well, who knows? It’s better than nothing, I suppose. Although, since I have never once required the services of a modprobe equivalent on Windows, I’m not even sure about that.
    If you require todo equal to a modprobe tweak of driver settings under windows it reboot computer. Driver command line equal options under windows are in registry options. In fact you are required todo this to make some USB keys work under Windows. Basically you have been lucky DrLoser.

    http://www.thewindowsclub.com/usb-devices-not-working-windows
    Windows 8.1 users have been required to do it a lot more often DrLoser. I guess either you have thrown out functional flash drives or don’t have Windows 8.1 yet.

    DrLoser really you did mention something that is unique to Linux. Is hardware firmware and hardware drivers being split into two different items. This does have some very major practical advantages.

    Also DrLoser another case of comment on a topic you have no clue on.
    http://www.la-samhna.de/library/rootkits/detect.html
    LKM rootkits are in fact highly detectable. Only reason Linux will have a LKM rootkit and it is not detect is if the administrator has not run any validated audit tools.

    Rootkits in kernel space depend on creating abnormal behaviors. Yes Linux audit tools methods would detect kernel level rootkits windows driver signing lets slide.

    Yes deployment QA include running rootkit hunter programs that audit the kernel space on Linux systems.

    Some of those old rootkits cannot function any more because kmem does not exist. Yes everything hotpatched in the Linux kernel is detectable.

  8. Deaf Spy says:

    Pogson, your only argument is that Debian has an “excellent repository”. Well, nice for Debian. Doesn’t seem to have been working for it last 10 years or so. Now, why should it work today?

  9. DrLoser says:

    DrLoser wrote, “What “key software innovations” has Linux “developed?””

    Loadable Kernel Modules (Linux v 1.2) …

    That’s hardly a “key software innovation,” Robert. It’s a “key kernel innovation,” maybe. Or possibly a “key firmware innovation.”

    And that’s assuming that it’s “key,” which it isn’t.

    And also assuming that Linux got there first, which it didn’t.

    How about Netware Loadable Modules in 1991?

    I’d imagine there is other prior art. I’d look it up, but what’s the point? The “key” features of LKMs are:

    1) The kernel intentionally has no stable API or ABI, so they’re pretty much borked from the start.
    2) LKMs offer a really neat attack vector into the Linux kernel. Yes, you do have to have elevated privileges in the first place, but LKMs offer a lovely way of dynamically hiding bits of rootkit once you’re there.
    3) There’s absolutely no QA whatsoever for an LKM, which is especially wonderful when you consider the lack of stable APIs and ABIs. Works on this point release? Great. It may or may not work on the next one. Or it might just look like it’s working, and crap on the system.

    Well, who knows? It’s better than nothing, I suppose. Although, since I have never once required the services of a modprobe equivalent on Windows, I’m not even sure about that.

    Exciting as LKMs are, Robert, they solve a Linux-specific problem in the usual half-baked Linux-specific way.

    Let’s hear about a key software innovation that users actually care about, please.

  10. olderman wrote, “so what is your point?”

    My point is that Deaf Spy wrote, “Windows 10 is rumored to include desktop apps in the Store, too. Store apps update automatically, there is a working review and rating system. What is the advantage, again?”

    I replied that GNU/Linux is here now and it’s not vapourware. You guys are like dogs fighting over bones. You make no sense sometimes.

  11. luvr says:

    “there will be no Windows 9. Microsoft is going from 8.1 straight to Windows 10.”

    Thanks for the info. Still makes me wonder what happened to Windows 9, though.
    Perhaps they took “8.1” and decided to add the “8” and the “1” together, to arrive at “9”… Thus, they already had a Windows “9”.

  12. Deaf Spy says:

    GNU/Linux is here now. It’s not vapourware.

    So is Windows 8.1 with Bing. Third time: what is the real advantage of using Linux for these devices?

  13. oiaohm says:

    luvr there will be no Windows 9. Microsoft is going from 8.1 straight to Windows 10.

  14. luvr says:

    “Windows 8 has it’s Store. Windows 10 is rumored to include desktop apps in the Store, too.”

    With all that talk about Windows 8 and 10, I’ve been wondering lately what happened to Windows 9. Wasn’t that supposed to be the next Windows release? Nobody seems to talk about it any longer. Has it been cancelled?

  15. olderman says:

    “GNU/Linux is here now. It’s not vapourware.”

    So is windows, and more importantly so are the applications that run on it.
    so what is your point?

  16. olderman says:

    “Uh, the next billion users of PCs may well choose to buy or not based on that price. ”

    But this discussion is not about the next billion users, it is about current users.

    so what say you, Robert Pogson?

  17. Deaf Spy wrote, “What is the advantage, again?”

    GNU/Linux is here now. It’s not vapourware.

  18. Deaf Spy says:

    Pogson, you are moving the goal post.

    We speak of small, cheap devices. Devices below the $200 threshold. Well, let’s even make it below $150.

    The only advantage you mention is the Repository. Windows 8 has it’s Store. Windows 10 is rumored to include desktop apps in the Store, too. Store apps update automatically, there is a working review and rating system. What is the advantage, again?

  19. olderman wrote, ” DO you really believe Robert Pogson, that the average non geek computer user who is fluent in working in the windows environment with windows applications is going to throw all of that away to save $100.00 on a new computer.”

    Uh, the next billion users of PCs may well choose to buy or not based on that price. See smart thingies. The same could apply to young/poor people anywhere. Even if $100 is a trivial amount, it could be better spent on other things, like electricity to run the computer for quite a while or food or Internet connectivity. There are many who pay M$ $hundreds per PC, OS and office suite included so $100 is at the low end.

  20. olderman says:

    “There are two cases:”

    Neither of which has any meaning to the average computer user. DO you really believe Robert Pogson, that the average non geek computer user who is fluent in working in the windows environment with windows applications is going to throw all of that away to save $100.00 on a new computer.

    Set aside your ideological baloney and think about it.

  21. Deaf Spy wrote, “My point is slightly different. Why bother learning a new OS with new window manager and take the risk to face the lack of applications, when you can get what you already know, with full application support, for the same amount of money?”

    There are two cases:

    1. Ordinarily priced devices on which M$ collects huge licensing fees. There GNU/Linux clearly has price-advantage by ~$100 or more depending on version and applications.
    2. The small cheap computers on which M$ is apparently donating licences for $0 or actually paying OEMs to install that other OS. There GNU/Linux still has the other advantages of flexibility because there is no restrictive EULA and the licence allows copying, using, modifying… Then there’s malware and re-re-reboots, and …

    Either way, in a competitive market, GNU/Linux will take a good share of devices. Debian GNU/Linux has a further advantage, the wonderful APT packaging tool and the depositories of Debian. These make managing one or thousands of devices pretty simple, even scriptable.

  22. Deaf Spy says:

    but they are wrong to claim it must always have that other OS on it.
    I’ve never made such a claim. My point is slightly different. Why bother learning a new OS with new window manager and take the risk to face the lack of applications, when you can get what you already know, with full application support, for the same amount of money?

    China and India are producing dirt-cheap Windows tablets these days. You have things priced < $100, powered by Atom CPUs.

    Really, what is the actual advantage of Debian or Ubuntu on these devices?

  23. DrLoser wrote, “What “key software innovations” has Linux “developed?””

    Loadable Kernel Modules (Linux v 1.2)
    From linux-1.2.0.tar.bz2: “Jan 31 1995 README.modules” This predates that other OS by months. This feature is sweet in GNU/Linux. It allows one build of the OS to run on rather diverse hardware because devices/drivers are matched at boot-time. Even as late as 2006, I was seeing that other OS freeze with the simple sabotage by students of swapping mice…

  24. DrLoser says:

    Interesting (though unavoidably ignorant) theory, though, Dougie.

    M$ has never developed a key software innovation and is not that good at predictions.

    You may very well be correct, allowing for weasel words like “develop” and “key.”

    Just out of mild interest, and I should be asking an expert like Robert rather than a mere salesman like you …

    What “key software innovations” has Linux “developed?”

  25. DrLoser says:

    Loser purports to have worked on BING, but cannot use a search engine to reference a cite..perhaps Loser is Google-agnostic?

    Loser has tried both Bing and Google, yea unto the third page or so, and Loser has found not a single piece of supporting evidence.

    Loser quote not did.

    Dougie he man big big assumption requires.

    Dougie big man he big man need big up.

    Ya up to it, ya ignorant little freak?

  26. DrLoser says:

    The exact line segment from 1981: No one will need more than 637 kb of memory for a personal computer

    I almost prefer Dougie’s imaginary version.

    It’s marginally less insane.

  27. oiaohm says:

    The exact line segment from 1981
    No one will need more than 637 kb of memory for a personal computer
    Remember is 1981. 1981 there are Workstations and Personal computers. Also that was also said only in the context of 1981.

    Dougman yes you are quoting a myth. The real line from Bill Gates is just as stupid since quote it is also quote it out of context. Notice its not a complete line the next word in the line is today.

    No one will need more than 637 kb of memory for a personal computer today. Yes perfectly correct for 1981 and it this line that evolve.

    “640K ought to be enough for anyone” is in fact the journalist head line on the article covering Bill Gates talk where he said that line because even back then it was believed to be stupid and the journalist had not got the complete quote either. Yes sometime you end up associated with the headline not what you said and sometimes it stuff take out of context.

    Yes DrLoser this is why I have yelled at you guys from time to time for not taken complete context parts.

    Some people think they have found a source document containing the quote so kept it alive completely missing its a journalist written title line. Just to be a problem the journalist responsible was not writing under there own name and the paper its from records are no more. So really “640K ought to be enough for anyone” by annoymous in 1981 would be very correct. It was written in 1981 we just don’t know by who.

  28. dougman says:

    Loser purports to have worked on BING, but cannot use a search engine to reference a cite..perhaps Loser is Google-agnostic?

    We are getting off topic, but here is another gem from the mind of Billy.

    “Prediction: Gates’s 286-page book mentions the World Wide Web on only four of its pages, and portrays the Internet as a subset of a much a larger “Information Superhighway.”

    Verdict: Miss. Gates’s notion that the Internet would play a supporting role in the information highway of the future, rather than being the highway itself, was out-of-date the day The Road Ahead was published. Even Gates realized it. Shortly before his book hit the stores, Gates reorganized Microsoft to focus more on the Internet, and he made major revisions to a second edition of The Road Ahead, adding material that highlighted the significance of the Internet. In many ways, Gates’s cloudy crystal ball regarding the Internet amounted to wishful thinking.

    M$ has never developed a key software innovation and is not that good at predictions.

    I searched Google with the terms “Microsoft innovation” to try to prove myself wrong but I did not find anything. Try it for yourself…I dare you.

    The best innovation from Microsoft I could think of is DOS, but it was originally written to IBM specs then Microsoft recycled it into MS-DOS which is more a profiting after the fact attitude.

  29. DrLoser says:

    You’re not really any good at this game, are you, Dougie?

    Apart from anything else, it’s trivially easy to see when you feel insecure and challenged, because you don’t include the usual plethora of links.

    Have you considered, say, signing up for a university course or two?

  30. DrLoser says:

    Even Billy said, “640K ought to be enough for anyone”…then tried to say he did not… so you see …

    … that “even a knowledgeable person like you, Dougie, is not averse to perpetuating myths?

    Go on, give us the cite for the quote.

  31. dougman says:

    Re: “Seriously, I still don’t see the market for these things”… just like you don’t see a market for a $99 Win-Dohs tablet? People like yourself, remind me of others that have come before you, myopic and narrow-minded.

    Even Billy said, “640K ought to be enough for anyone”…then tried to say he did not… so you see, even a knowledgeable person such as him, does not know what the world needs to know or have. Even Jobs said, that people don’t know what they want, but Chromebooks have hit a sweet-spot for simplistic computer anyone can use and not have to suffer the woes of Win-Dohs.

  32. DrLoser says:

    (Also, by “manufacturer,” I mean “OS vendor.” Must be short of caffeine today. Apologies again.)

  33. DrLoser says:

    (64GB of HDD, dammit! Apologies.)

  34. DrLoser says:

    Further, a ChromeBook doesn’t expire when M$ says the software expires.

    Who knows, Robert? The things have only been in mass circulation for about three years. If you are correct, and people simply brick over them with the Distro of Choice, then their entire point (from Google’s perspective) disappears. Even side-loading an alternative browser or … the ignominy! … an AdBlock script would render them pointless as far as the manufacturer is concerned.

    OTOH, they might very well last ten or more years. Who’s to say? Data on how many of the early models are still in use would be very interesting here, don’t you think? (Honest enquiry.)

    If I was going to brick a Chromebook and install that Distro of Choice, btw, I think I’d want a darned sight more than 64GB of RAM. Maybe I’d go for something like the Acer C7, which has a more reasonable 320GB. At £250 brand new.

    Or maybe I’d just buy a cheap notebook (with or without Windows) and save myself the hassle.

    Seriously, I still don’t see the market for these things, Robert. People are going to be sucked in by the price and repelled by the “feature set.” It actually looks like a bad marketing move by Google.

    But that’s just an opinion — very much like yours. It’ll certainly be interesting to see what the next couple of years bring.

  35. dougman wrote, “Chromebook and Samsung S4, ~5% of the time I have to set at my desk.”

    I think Deaf Spy and others have a point that a desk with a legacy PC on it may be needed for heavy typing, scanning, printing, whatever… but they are wrong to claim it must always have that other OS on it. GNU/Linux works everywhere I’ve been except in a tiny proportion of seats where some force majeur demanded some particular locked-in application. That may apply in the main office and on a few workers’ desks but certainly not every desk. There are workarounds too, like RDP so that GNU/Linux does not even have to be on the desktop machine where a user is using that other OS and some application. Even folks who use that other OS don’t have every application on that one desktop. That would be just a silly waste of resources. Munich found hundreds of applications in their system and they rationalized them down to just a few PCs using that other OS.

  36. dougman says:

    Re: “but sooner or later they come to the need to do some, gasp, real work.”

    Yes, say like, reinstall the stupid OS due it to being pwned by malware.

    I don’t know about you, but I do majority of my work from my Chromebook and Samsung S4, ~5% of the time I have to set at my desk.

  37. Deaf Spy wrote, “sooner or later they come to the need to do some, gasp, real work.”

    My Little Woman does some real work but all except a few documents are on the web. She could do very well with Google Docs. When LibreOffice has a web interface, she’s good with a ChromeBook. At the moment she has the browser, the file-manager and a weather-plugin running, no other user-applications. She would be better off using a server with document/image databases than that file-manager and the weather can be followed on many web-sites. She has a dual-core 64bit processor running at 2.2gHz and 1gB RAM, easily within the specs of small cheap computers. Her machine is ancient, 2007/8, and is probably due for an upgrade soon. A good ChromeBook might be the ticket or we could upgrade her software or RAM for a bit more life. She’s running 32-bit software because OpenOffice.org wasn’t 64-bit back in those days but LibreOffice works fine today. She might get a boost in performance that way but her machine is noticeably faster than a couple of Atomic systems we have. Maybe that’s a project for the coming year. She does need a new sound system. The USB-speakers she has been using died.

  38. Deaf Spy says:

    A huge fraction of humanity does nothing with a PC except browse the web and move snaps of the family around FaceBook
    This is a misconception of greatest proportion, Pogson. People may live well browsing the web and Facebook, but sooner or later they come to the need to do some, gasp, real work.

    Speaking of cheap, see what’s happening in the wild wild east:
    http://www.infoworld.com/article/2825878/mobile-technology/cheap-windows-8-1-tablets-rush-to-the-bottom-windows-rt-loses-big-time.html

    Plug in an external keyboard and display, and voila! The small cheap computer you’ve been dreaming of! Only this time with full set of real, productive applications along. Nice, isn’t it?

  39. Deaf Spy wrote, “I would stick with the ChromeBooks
    Crippling yourself consciously”

    Nope. If I don’t need/want to run any .exe, I don’t need that other OS. I am not crippled at all. A huge fraction of humanity does nothing with a PC except browse the web and move snaps of the family around FaceBook. A ChromeBook will do that just fine. I do like a more general purpose OS. That’s why I use GNU/Linux. Consumers can too. GNU/Linux is not cripple-ware and it doesn’t need all that other trash that other OS requires: anti-malware, registry delousing, re-re-rebooting, almost auto-updates, etc. Further, a ChromeBook doesn’t expire when M$ says the software expires. The ease of maintenance is just what a lot of folks like about ChromeBooks. That other OS is about lock-in the oppposite of ease of maintenance.

    There is such a thing as slavery to M$. Just read M$’s EULA.

  40. Deaf Spy says:

    Let’s be clear: A Windows notebook can do everything (with a few important exceptions) that a Chromebook can do. The reverse is definitely not true.
    versus
    I would stick with the ChromeBooks

    Crippling yourself consciously… Masochism, zealotry or… Perhaps slavery to ideas?

Leave a Reply