jaclaz

The point I was trying to make is slightly different. Of course the good MS guys have all the rights in the world (+one) to change the way (be it protocol/urls/deployment mechanism/whatever) they deliver updates and they are perfectly free to do so twice a week (if they like to) and as well to refuse updating an EOS Operating System, etc. I would go even further, they don't have any obligation whatever to explain what any update does, for all it matters they could publish updates with this accompanying documentation (fake): A non-explanation is IMHO better than a vague (and in this particular case meaningless) one. Now, IF they do provide an explanation/documentation for an update I expect it to be: clear accurate truthful and would like it to be also: 4. exhaustive Additionally in this specific case of unspecified improvements to the "overall client", it has to be noted how after all it is not brain surgery or rocket science, the good MS guys make a list of new updates and publish it on their site, then the Windows Update client looks periodically at that list and downloads what is considered "necessary" with - maybe - a minimal amount of checking when a same update may be compatible or incompatible with the actual machine at hand. The procedure is so simple that it either works or it does not, binary 0/1, Off/On. IF they change anything in the way the updates are delivered then the "old" client version will stop working and they will need to deliver (through the previous protocol or as a separate manual download) the "new" client compatible with the new way/protocol/etc. jaclaz

Well that's the typical effect of the continuous update process (as MS has planned and implemented it), they wait for you (or the computer) to go to sleep and then they do to your poor, unsuspecting system whatever they fancy. Every day you will boot to a new OS, with a number of changes done overnight. (for your own good of course) jaclaz

Well, the issue is with the definition of "improvements in the overall Windows Update client" which doesn't really mean anything. Will the Windows Update client (once those two Kb's are downloaded) be: faster more reliable (which would imply that all the people that don't use those two kb are still successfully having updates but in a less reliable way) have a priority in the downloading queue on the server (if such thing exists ) allow for no (or at least less) "mandatory reboots" after the install of the downloaded updates occupy less (NOT more) bytes on the system disk deliver/download bolder 1 and fatter 0's If one (or more) of the listed items is true, then we may talk of specific improvements (and these may be compared with "overall improvements" which still make some sense, but still not with "improvements in the overall client" that make no sense whatsoever, at least linguistically). jaclaz

And here is Tesla : http://www.wired.com/2015/08/researchers-hacked-model-s-teslas-already/ jaclaz

I wouldn't be surprised if there was a service running a .Net assembly/container relying on a XAML file to implement a dual mode Flash/HTML5 timer/ticker (that can only be rendered properly by Edge or IE11), with values obviously expressed in nanoseconds in order to make leverage of 64 bit numbers . jaclaz

Well, the good MS guys must have also invented a time machine. At least at end of june 2015 submissions were closed: https://web.archive.org/web/20150627183454/http://www.idsa.org/awards/idsa-awards it is doubtful that the jury can have actually tested the final version. The award is for the "design" not for the implementation , very likely the jury has seen some mock-ups and maybe some half-@§§ed prototypes, it is a (anyway perverted) appreciation of the looks of it (while obviously the only things that actually counts is functionality of the whole OS, not just of the Start Menu). Check the other awards: http://www.idsa.org/awards/idea/preview The Digital Design award given to Microsoft has the same dignity and relevance as - say - the Packaging & Graphics one given to Purina for their Brand New Cat Litter: http://www.idsa.org/awards/idea/purina-r-pro-planr-renewtm-cat-litter-jug And even if we remain into the "Digital Design" category, there are several "Top Winners" (right now they are Gold and Silver mixed together), besides MS: http://www.idsa.org/awards/idea/digital-design/humax http://www.idsa.org/awards/idea/digital-design/naver-service-history-wall-connect-one http://www.idsa.org/awards/idea/digital-design/post-it®-plus-app http://www.idsa.org/awards/idea/digital-design/sejong-city-glass-wall http://www.idsa.org/awards/idea/digital-design/tomra-cui http://www.idsa.org/awards/idea/digital-design/attractive-concierge http://www.idsa.org/awards/idea/digital-design/dialogue-dark On the 22nd of August when the actually awarded the Gold Award will be revealed, we'll see if they are actually convinced that the Start Menu actually deserves it. Anyway, and with all due respect to the good guys that assign the awards the whole thing more than a competition seems a lot like a showcase for industrial products (or prototypes) that are "new" and possibly "weirdly" designed or exploring the borders of - well - *something*, example jaclaz

Well, you could still use a self-deleting batch in a SFX archive. Loosely, connect the dots between this (seemingly unrelated): http://www.911cd.net/forums//index.php?showtopic=21123 and this: http://www.catch22.net/tuts/self-deleting-executables jaclaz

And unsurprisingly.... http://www.computerworld.com/article/2954668/telematics/hacker-shows-he-can-locate-unlock-and-remote-start-gm-vehicles.html jaclaz

Now, on "other news" interesting "terms of use": https://hulbee.com/agreement (that could be well titled "What Google does ...") and "privacy statement": https://hulbee.com/privacy From this newish (Swiss) search engine: https://hulbee.com/ I particularly like the part between brackets and the "only-if-forced-to" policy, as opposed to the traditional "opt-in" or "opt-out" (which I made BOLD RED to highlight): jaclaz

Can you provide some details on how you got it? Going to your link http://gs.statcounter.com/#desktop-os-ww-daily-20150701-20150805 I can find not a Windows 10 at all. besides the usual suspects: XP/Vista/7/8/8.1 I can find only OSX, Linux, "unknown" and "Other" jaclaz P.S.: EDIT, I need to set the date range within 29th July - 4th August to have Windows 10 "magically" appear in the legend and graph, here is the link: http://gs.statcounter.com/#desktop-os-ww-daily-20150729-20150805 the different scale somehow highlights a hiccup in 7 stats centered on the 2nd of August and a return to "normal levels" on the 3rd and 4th (it must mean something .... ).

Not exactly-exactly, the screenshot: says "only available for a month after" for all we know may mean that it "disappears" or that the text will remain as is and that the Get Started button get disabled or that the text changes to: or that everything remains as is and when you press the "Get Started" button you get (say) a Blue Screen Error or any "normal" (in the sense of normally clear) Windows error like "Something happened". And one month is 30 or 31 days (or 28 if you installed Windows 10 on 1st of February)? Exactly. jaclaz

I guess that we could use comparative terms . XAML is as smart as .Net XAML is as useful as Silverlight .... BAML is as stupid as .... (no wait, I don't think there can be something stupid enough to be compared to it ) jaclaz

I don't know , IMHO the thingy is too new for making (yet) this kind of statements: Surely this is what Windows 10 says , but when a month will have passed and someone will actually try to upgrade back to Windows 7 we will know what actually happens. There are also some legal implications in the stuff as I see it. The License for a previous OS version is to all effects a "lifetime license", I find it possible, if not probable that someone that will have done the (free but somehow "forced") switch to windows 10 may decide to go back after more than 1 month and if this will not be allowed will take some form of legal action against MS. On the other hand, don't call me a pessimist, I am just thinking out loud, there will be millions of "original", "legitimate", "COA" (and what not) Windows 7 DVD's around that will arrive to the second hand market that if used for the switch to Windows 10 will not work/activate and a very few that (since they were not used for the switch to Windows 10) will be - at least until EOS in January 14, 2020 - perfectly valid with NO WAY to know which is which until install/activation time. jaclaz

Not exactly-exactly. DOS and Windows 9x/Me need to be on the First Active Primary partition of First Disk (that will be assigned drive letter C: at boot time automatically, the difference is subtle but important). This same (Active) partition must also be (unless a third party bootmanager is used) the NT system "boot" partition (what Microsoft calls - reversed - "system") while the actual operating system can be on any volume (that can be either another primary partition or a logical volume inside extended) on any disk (not necessarily first disk) this is what anyone would call "system" volume (but that the good MS guys call "boot"). About boot/system naming: http://www.multibooters.co.uk/system.html And more generally about dual booting a DOS and a NT: http://www.msfn.org/board/topic/169320-dual-booting-dos-and-win7/ jaclaz

Well, to be on the safe side (since you Czech version may have some twists) you could use the more "traditional" F6 floppy approach. That is the "safest" method and is used (by means of a virtual floppy created in grub4dos) in a few of the "Install from USB" methods, it's just a matter of porting to CD the "from USB approach", or - if you prefer have the original (untouched) XP .iso inside a grub4dos bootable .iso. Particularly this "base" method: http://www.msfn.org/board/topic/149675-install-xp-from-a-iso-file/ should be easy to adapt to this different way (bootable CD as opposed to bootable USB device), you just need a floppy image containing both the Winvblock driver and your drivers, something that has been also already done and tested (on USB), like here: http://www.rmprepusb.com/tutorials/install-xp-from-an-iso http://reboot.pro/topic/18654-how-can-i-load-more-than-one-driver-using-xp-txtsetupoem/ http://www.rmprepusb.com/tutorials/72---easyboot---a-grubdos-multiboot-drive-that-is-easy-to-maintain/e2bv1/dpms_srs There shouldn't be particular issues with adapting the thingy to work "from booted CD" as opposed to "from booted USB" and it may even work in CD Hard Disk emulation mode , but really-really can't you use a USB stick like all the rest of the people nowadays? Besides being faster than CD/DVD it is much easier to make experiments/changes/whatever instead of re-burning optical media though a large part of the tests might be made in a VM when it comes to installing n OS it is not rare that a VM behaves differently from a "real" machine. As a side note, if the BIOS of your motherboard has "IDE compatibility mode" or similar settings in BIOS, you can always post-install the SATA driver, see: http://www.msfn.org/board/topic/120444-how-to-install-windows-from-usb-winsetupfromusb-with-gui/page-24#entry884409 http://forums.pcper.com/showthread.php?444831-HOWTO-enable-AHCI-mode-after-installing-Windows also to be "tuned" for your drivers/hardware but all in all doable. jaclaz

Yes. http://homepage.ntlworld.com/jonathan.deboynepollard/FGA/questions-with-yes-or-no-answers.html http://unattended.msfn.org/unattended.xp/view/web/33/ http://unattended.msfn.org/unattended.xp/view/web/36/ http://ryan.servehttp.com/tech/windows/howto_xp_slipstream.htm jaclaz

Well, whenever someone (anyone) gives you numbers or statistics, a good idea would be to check and double check them, sometimes they are simply wrong or misrepresented or misinterpreted. Right now on Wikipedia the netmarketshare data for Windows 10: https://en.wikipedia.org/wiki/Usage_share_of_operating_systems is at 2.47% while the original data on Netmarketshare: http://www.netmarketshare.com/operating-system-market-share.aspx?qprid=10&qpcustomb=&qpcustomd=0&qpsp=198&qpnp=1&qptimeframe=M is at 0.39% As a matter of fact if you sum the whole column on Netmarketshare it's result is 100.01, while the whole column on the Wikipedia page is 102.00 Of course in this case it is most probably a slip of the finger/typo, and has no practical consequences, but the same thing (or a similar one) can have really dramatic consequences, last world wide example being the (in)famous Reinhart-Rogoff Excel mishap: http://theconversation.com/the-reinhart-rogoff-error-or-how-not-to-excel-at-economics-13646 a number of countries adopted (completely wrong) economical politics based on an already debatable enough theory based on false/wrong data. jaclaz

Wild guess , mind you, but it is possible that those new sticks are set as "Fixed" media and this *somehow* affects the behaviour of the booting PE. Basically 99.99% of USB 2.0 sticks are set in factory as "Removable". Some of the fastish USB 3.0 are not anymore a USB stick, but rather a USB to SATA bridge with attached to it a SATA (miniaturized) SSD and they thus come as "Fixed". Since all in all the drive letter is linked to the volume GUID in DosDevices and a hard disk like device (Fixed) uses disk signature+offset as identifier while a Removable device uses a different method, it is possible that the issue revolves around this. Another reason might be if (for whatever cause) the wpeinit.exe is not executed. Try running: wpeutil updatebootinfo https://technet.microsoft.com/en-us/library/dd744592(v=ws.10).aspx And see if it makes a difference. In any case you can check in the batch (which contains NOT "DOS commands" but rather Windows Command Line ones) for the presence of a "tag" file in the available drives and set your variable with the drive letter resulting from the check. Not entirely unlike we did in the good ol' times: http://www.msfn.org/board/topic/137714-install-xp-from-a-ram-loaded-iso-image/?p=895361 http://www.msfn.org/board/topic/120444-how-to-install-windows-from-usb-winsetupfromusb-with-gui/page-72#entry1005836 jaclaz

Not really, most probably there is also data for when you leave the machine unattended for a few minutes (to get a cup of coffe or smoke a cig), one of the points of telemetry (as implemented by MS, and set aside the privacy concerns) is that you have no access to it's data, and the good MS guys (particularly the PR and marketing guys) can say anything about it, just like statistics (when results are published without providing the RAW data). http://thinkexist.com/quotation/oh-people_can_come_up_with_statistics_to_prove/338856.html I will go further in the past a number of boasted numbers about SALES of MS (OS or hardware) were later debunked by analyzing quarterly or yearly financial reports, in this case, when there is no sale as the upgrade downgrade is free, they can really tell you *anything* (and the contrary of it). jaclaz

Well, with all due respect , allow me to doubt this without some more data/reports/anaysis. At least historically, the good MS guys have always made things more complex (often without actually any need for it) than what they were in a previous version, so they already had the "voting system" in earlier OS's, like: http://www.msfn.org/board/topic/128378-hardware-changes-and-windows-activation/?p=827471 it would be really news if for once they made something actually simple and straightforward (like "you can change anything but the motherboard"). jaclaz

And it's only fair to link to someone which has different views on the thingy: http://www.boycottdocker.org/ Whom actually, while comparing the pro and cons of Docker containers against VM's (Virtual Machines) and criticizing most of the design (or of the hype around it) writes seemingly in plain enough English thus allowing us "normal people" to get an idea of what it is. jaclaz

I'll throw this on the table (and promptly hide my hand behind my back ): http://tinywall.pados.hu/ if one has anyway the stupid bloat .Net is, it is actually lightweight and represent IMHO a more friendly way to manage the Windows built-in firewall. As a side note, the views of the Author on security not necessarily being the same as mine BTW, I would give them a 70% hit rate: http://tinywall.pados.hu/sectips.php though Tip #10 is obviously a very, very good one, and compensates for the little sense that #1, #2 and #8 make. jaclaz

Most probably (but cannot really say ) nothing has changed in Windows 10 when compared with 8/8.1 so making a full trace like explained here: http://www.msfn.org/board/topic/158252-trace-why-windows-8-boots-shutsdown-or-hibernates-slowly/ may apply as well (to find out the culprit). jaclaz

Hopefully Of course it is just fun but I like to imagine the stupid Windows 10 much like an automated Mechanical Turk , with a little girl, Cortana, inside it actually reading everything, interpreting commands etc. and then doing whatever things she fancies, not necessarily what the user asked for .... jaclaz

JFYI, 7-zip has a built-in feature to calculate a number of common hashes, including CRC32. jaclaz