Drugwash

We still run 9x in 2016 with no support from "mommy" Microsoft, why not run other "unsupported" software as well? Remember: Tihiy by himself years ago created LameSkin which evolved into Revolutions Pack while everyone around mocked him saying it can't be done. We should prove Microsoft and anyone else in their stronghold that we have better programmers and they are not for sale!

Something like this? http://www.theregister.co.uk/2016/01/25/linux_foundation_scraps_individual_membership/ Not exactly that one but along the same lines.

How about going at 0x0A31 and replacing 0x75 with 0xEB? That'd be less error prone. EDIT : For the sake of completeness, let's put it this way: - the portable version of Palemoon is compressed with UPX so it has to be uncompressed before patching i.e. upx.exe -d drive:\path\to\palemoon.exe - at hex address 0x0A31 replace value 0x75 with 0xEB (that'd be JNZ --> JMP) any hex editor would do, I use XVI32 just because I had it around - save the edited file in my tests it didn't need any certificate removal or checksum fixed - compress the executable back using UPX (optional, it runs fine as it is) i.e. upx.exe --best drive:\path\to\palemoon.exe

Linux has already been compromised by M$, Google and others (there was a recent link around here exposing a few things).

If that guy were a real programmer he would've put a one-time messagebox stating something like "You're a stupid user using a stupid hack, therefore don't come crying to me when your computer blows up because you ran my software. Do you agree to this? if not, eff off!" and a couple buttons labelled 'Yes' and 'No'. The choice belongs to the user. I was about to say more but I've had enough barking up the wrong tree. Real programmers died long ago and a new generation might take too long to rise again.

Unfortunately they're not playing a fun game. When there is no choice we must create one. Forcibly, if there's no other way. But a clever one at the same time.

The PCI slots are identical in shape but there is an underlying software difference. Unfortunately, the so-called search engines don't care about search terms and just throw at the user whatever garbage has been more "requested", therefore I couldn't find an article describing differences between PCI (and not d4mn PCI-express) versions. However, there may be little chances that such difference may affect the functioning of a videocard but please do keep it in the back of your mind, just in case. I have no idea if the dreaded spying software known as Windows 10 will recognize and enable a PCI video card, but if you set the BIOS correctly (as I already said somewhere above in regard to slot priority, providing that machine's BIOS does have such feature) you may at least see the POST screen while the computer starts, if only for a split second. Actually, that's what you should've done with your current videocard already, to verify if it's in good shape or not. There's no point in acquiring a bunch of videocards if the Deskpro simply doesn't accept a PCI videocard. I assume you already know this, but just for the sake of it: if the card doesn't first fit visually in the slot, do not attempt to force it in!

Some of us, very few, need another planet to live on. There is no alternative to that other than death.

The stupid check is in the sources at browser\app\nsBrowserApp.cpp (I got the Atom/XP sources): #ifdef XP_WIN HKEY hKey; LONG lRes = RegOpenKeyExW(HKEY_LOCAL_MACHINE, L"SYSTEM\\WPA\\PosReady", 0, KEY_READ, &hKey); if (lRes == ERROR_SUCCESS) { Output("Unsupported operating system."); return 255; }#endifThing is, heinoganda's patch may be quick and dirty but it only works for this version. Actually I haven't tested it on the portable version I had unpacked, which is UPXed. We need a maintainer that can rebuild each new version with that check disabled and maybe tweak something more out of it, if needed. Or just forget PaleMoon ever existed.

Ironically, I had a slightly older version of PaleMoon - 25.3.1 - installed on XP but due to certain bugs I haven't used it for more than a few tests at the time. Then, a month or two ago, I picked it up again, updated it to 25.8.1 when it got released, installed a few more add-ons (even modded a couple to get them running properly) and found it was actually a bit faster and less resource-hungry than the previously used Firefox 13.0.1. And now, just when I got used to it and was almost perfectly set up, there they go: new engine, dumped compatibility. My question is: do they do this willingly (I mean all developers from major software projects) or is there something behind the scenes? Because real programmers could always work their way around such (non)issues. In regard to PaleMoon, at least the source code is publically available but who has - or can build - a proper development environment to patch the code and release POSReady-compatible versions? Unfortunately I can't. Anyway, I'm about to unpack the portable version of the Atom/XP v26.0 in a separate folder, duplicate the existing profile and see if it's gonna work on my XP-SP4 machine, because the unofficial SP4 does create that POSReady registry key (just checked and it's there). EDIT : Not working. So I saved myself HDD space and frustration by deleting ~24MB of "new" PaleMoon crap. Current version still works, I'll use that one for as long as possible. And then I'll switch over to KolibriOS.

There will come a time when they won't hide anymore.

OK, now I'm officially craving for fried fish. No, I'm not pregnant!

According to the link in post #1285, freedom - as in 'open-source' at least but not only - is just a word in a dictionary (still, we don't know for how long). I've said before that 'there can be only one' and that M$ might soon buy out Apple. It may sound crazy, utopian, but the way things go this is the only logical course of action. They don't want clever/smart people, they don't want people that can say 'NO' firmly and knowledgeably to their bull$h!t. They want full control and they'll get it. Sadly, they will. I just hope I won't be around by then. Unless the world gets rid of them M$, Google, Apple and all other huge corporations that hold monopoly to everything, soon including our own thoughts. Yeah, utopian too.

Keep looking for a video priority setting as per my comment #5 above. If that's not available then all you can do is test the videocard on another (working) computer to check if it's not damaged or try a different (presumably working) PCI videocard on this computer.

A better PSU (as in higher output power) can offer more current to any device that requires it, PCI devices included of course. Personally I stil believe either the BIOS does not acknowledge the PCI videocard, video priority is not correctly set or the card is defective, otherwise the machine would just shutdown shortly after power up or reboot constantly, if the PSU were too weak. But it's just an opinion that could be wrong. One way or another you have to prove through tests that each piece is in good shape and working correctly.

The human-readable description of the device is picked from the driver inf file. it's entirely possible that such file may contain an erroneous description that may have been fixed in a later version of the driver.

You can use a bootable CD/DVD with the appropriate tools to test RAM, HDD and other parts. MemTest86+ is a good RAM test. HDAT2 can test and repair bad sectors in HDDs. If you can't afford the repair shop you'll have to learn and do it yourself. Preferrably without breaking anything. If you're willing to get to the bottom of this in regard to the Deskpro, find a standard PSU somewhere around 230W or higher, reconfigure the wirings with extreme care (and triple-checking), connect it to the Deskpro's board and see if the new videocard works that way. Try other PCI slot, if available - there may be IRQ conflicts or even some defective slot that prevents it from working - assuming the card itself is in good shape. Please keep in mind that reconfiguring the PSU socket wirings is risky and can permanently damage the motherboard - don't blame me (or anyone else) if you blow it up! That said, good luck!

Linux is pushing updates just the same. I've already had such an update that broke and applet/desklet/whatever and editing such thing to make it work is a huge PITA, if at all possible. I can edit a Firefox addon to make it compatible with an out-of-range version or with Pale Moon, for example, but I couldn't get that Linux thingie to work. And to think it was a trivial non-issue such as an icon file (!!!) being present in a package that shouldn't (?) have been in. Unfortunately we live in a purely commercial world, not in a humane world, so nobody will ever build the perfect product because they think they'll immediately go out of business. And everything, just everything in the world is related to this kind of thinking, which in turn relates to money. It's always, ALWAYS about the money. Nothing more, nothing less.

I gave up TV and radio years ago - they're not only useless, they're dangerous through their manipulation of the truth in ways an honest person couldn't even imagine. Actually I think there's a law of the Universe that makes everything try to compensate to get to an even balance. Therefore the smartest TVs will get the dumbest viewers and the other way around.

Unless you have some tech-savvy friend you may be better off taking this computer to a repair shop, if you really want to see it working.

Unfortunately there is quite a distance and heaps of taxes between the US and Sweden (or Romania, for that matter) so the western market probably wouldn't help in this case.

They could be injecting some javascript or something else that reads file version of some key IE components. I noticed a couple days ago that User Agent Switcher by Chris Pederick doesn't work for Pale Moon 25.8 (Atom/XP version). It used to work for Firefox 13 though but I haven't used that one in a week or two. Have you tried to disable/enable active scripting in IE advanced properties? Maybe this has something to do with the issue. EDIT: After a bit of testing I notice failure in wininet -> InternetOpenUrl() when trying to reach the Mozilla Addons site with my script. I've even tried to send different user agent strins such as those pertaining to Firefox 40 and IE9, to no avail. Maybe they now require secure connections exclusively or there is something else I didn't figure out yet. Same function wrapper works in another script for a different site, so it's definitely related to particular sites. Same thing could happen to you in that game.

I could buy a used, fully operational Pentium III here for about $20. A 40GB IDE hard drive would be about $5. Other parts would be more expensive, of course, but not too much. Admittedly, second-hand computer market is slowly dying here too. @submix8c: judging by the wording in post #10 (doesn't take up much space) it appears to be a slim desktop case. But I may be wrong assuming this. If it indeed is a tower case then of course a regular PSU may be fitted even with modifications to the case if needed.

FlashGet is good (I'm using 1.9.6.1073 on both 98SE and XP) but occasionally it may behave. Have you tried wget? There should be a Win9x-compatible version. It's commandline-based and might sport the options you need.

Dunno if it's in any way related but recently one of my AHK scripts that was supposed to download certain elements off of Mozilla webpages started returning gibberish instead of HTML code and files. It's also based on the IE built-in API and therefore it's using either IE6 on 98SE or IE8 on XP-SP3. It used to work when I last tested it in october 2015. I never use Google but I just tried mail.yahoo.com in IE8 under XP and it brought me the login screen (I didn't go further). I have a feeling it may be related to whether the application calling the APIs is a known web browser or not.