Jump to content

Mathwiz

Member
 View Profile  See their activity

  • Posts

    1,851

  • Joined

  • Last visited

  • Days Won

    50

  • Donations

    0.00 USD 

  • Country

    United States

 Content Type 

Everything posted by Mathwiz

  1. Mathwiz
    Since this error appears to be related to missing Shadow DOM support, I'm testing in St 55.... BRB.... Well, that was sort of a bust. Couldn't really test anything since I'm not a Comcast customer (thank goodness), so no signing on for me! I'm guessing @Art7220 couldn't sign in with the pref disabled, got further with it enabled, but gave up because setting the pref broke YouTube (a la my experience with Chase). Be interesting for a Comcast customer to try this and see what the issues are with the pref disabled and with it enabled, not only in a UXP browser like NM but also in St 55.
  2. Mathwiz
    I had my suspicions; after all, it's intended to test whether or not a feature is implemented; it isn't an exhaustive test of whether a feature is implemented correctly. It's all I had to go on, though. So, St 52's implementation of Custom Elements is still incomplete. It's probably the best we have until Moonchild puts some effort into it, though. As for Shadow DOM, html5test.com claims St 52 has partial support, and that St 55 has full support. We'd probably need to scour Bugzilla for relevant code (and might, perhaps, find more relevant usable code in FF 54 and 55. I doubt anything later could be merged into either Serpent version though). BTW, it turns out that setting dom.webcomponents.enabled breaks Chase.com in St 52 too, although it's not as badly broken as in St 55. Apparently the site is designed to work correctly if someone disables Web Components, but otherwise it expects a complete implementation. So my advice now is to leave the pref off in both browsers. If anyone discovers a specific Web site that only works with the setting enabled, make a copy of your main profile, and just toggle that pref on in the copy, so you can use that profile for the site that requires it. (Let the rest of us know what you find too.)
  3. Mathwiz
    I don't think Mozilla got Web Components quite right until well into the Quantum era, but there is hope! Per HTML5test.com, latest St 52 has full Shadow DOM support (I'm guessing thanks to MCP) but lacks Custom Elements support, while St 55 has Custom Elements support but partial/broken Shadow DOM support, both probably inherited from FF 53. So perhaps we could port the changes for each into the other and get full Web Components support in both versions. Web Components are only one piece of the current JS issues, but it seems like it's worth a try.
  4. Mathwiz
    It's worse than that. I just found that setting either pref on Serpent 55 breaks chase.com. I'm guessing that chase.com checks whether these features are enabled and tries to use them if present, but runs fallback code if they're not. (Surprising when so many sites just expect them to be fully implemented.) If 55's Web Components implementation is incomplete and/or buggy, then enabling either pref actually makes things worse.
  5. Mathwiz
    On Serpent 55 enable that pref and dom.webcomponents.customelements.enabled too. This is probably the closest you can get to Web Components support on any of @roytam1's browsers (too bad 55 has fallen behind in other JS areas):
  6. Mathwiz
    Besides Notepad++ (which is great), Micro$oft's Wordpad, which comes with WinXP, works just fine. I use it all the time for these install.rdf fixes. Notepad will work too, but if the install.rdf has only LFs (not CR+LFs), then the lines of text will all be smushed together, making it very challenging to edit! Wordpad handles text files with only LFs much better. BTW, here's a trick I use: install 7-Zip, set up Wordpad (or Notepad++) as 7-Zip's default text editor, then use the "Open With" dialog to associate .xpi files with 7-Zip. Now you can open them by simply double-clicking, edit the install.rdf by clicking it and pressing F4, and when you close Wordpad and save your changes, 7-Zip will prompt you to update the install.rdf file inside your .xpi file automatically - no more need to extract, zip back up, and rename! Great time-saver.
  7. Mathwiz
    That one confused me until I looked it up on Wikipedia. Who in their right mind would encode a date in a format that only spanned the 22 years 2000-2021? Well, nobody, of course - so naturally Micro$oft did just that, in their Exchange email product! For us aficionados of older OSes, the "big one" is probably 2036, when NTP rolls over. I doubt even Windows 10 will get an update for that one! Luckily, we should all have a 3rd-party NTPv4 client by that time. The thing that annoys me about this one is, NTP uses a 64-bit time value, but splits it into two 32-bit parts: the first for the number of seconds since 1900 (what were they thinking?) and the second for fractions of a second, down to a resolution of 233 picoseconds, the time it takes light to travel about 7 cm. If they'd just started NTP at the GPS epoch, we'd have another 80 years, pushing the "crisis" up to 2116; if they had allocated even one more bit to the integer part (splitting it 33/31 vs. 32/32), we'd have another 136 years! But then light could travel a whole 14 cm before we noticed
  8. Mathwiz
    You did take note of my handle, didn't you? Hmm... that's the same point I made just a few posts ago - but was curtly shot down by.... I guess we'll just have to agree to disagree. I'm more interested in what Mozilla could have accomplished, had they stuck with UXP, or perhaps split FF into (ahem) "two distinctly different products" (say, FF Classic and FF Quantum); I guess you're more interested in things as they are now. To be fair to Mozilla, I understand that modern Seamonkey still retains a vestige of UXP support, although I haven't really followed SM much since the end of XP support. But SM isn't the product Mozilla pushes, and the SM user base is a small fraction of Mozilla's total. Exactly; that's what I was getting at. How long until we have to contend with FF 1000, even at the current rate? A very long time. But how long until there are 1000 different FF's floating around out there, when you include ESR versions and consider FF Android separately from FF Desktop? Not nearly as long. Heck, if you include Seamonkey and/or count "Nightly" releases, I think we already crossed the 1000 mark a long time ago!
  9. Mathwiz
    AIUI Mozilla is currently burning up version numbers at the ridiculous rate of 13 per year - more than 1 per month! But it'll still take them nearly 70 years to get to FF 1000. Of course, if they went to weekly updates, they'd get there in under 20. Even I might live to see that!
  10. Mathwiz
    How does he do it? Has he found - or created - a Rust compiler that targets XP? Could he let us (or at least @roytam1) in on the secret?
  11. Mathwiz
    UXP browsers like NM 28 are updated monthly now. Be patient. Besides, did you really feel the need to update your browser every week? Unless there's a critical security fix, even monthly seems like overkill to me. The Web doesn't change that fast! +1. It's just another instance of Mozilla chasing Google. Chrome is way up into the 90's, so Mozilla thinks FF has to be too. Are users really so dumb that they'd think Chrome 95 is "better" than FF 91 just because it has a higher version number? Well, I'm sure a few are, and they're both chasing the lowest common denominator.... I've mentioned this before, but it reminds me of when the DECT cordless phone standard came to the US. They named the US version DECT 6.0, even there was no version 1.0, 2.0, 3.0, 4.0, or 5.0 - because the previous generation of cordless phones worked on a 5.8 MHz frequency, and they thought folks would think "5.8" phones were "better" unless the new DECT phones carried a higher number Personally, I'd just use the year as a version number; you know, like Windows 95, 98, 2000 ... what was wrong with that? You could use the month as a subversion if desired.... Not sure what's going on here, but it's interesting to see updates to NM 27's Javascript engine.
  12. Mathwiz
    Thanks for that info; unfortunately, it makes the stats less useful IMO. Quantum is as different from UXP as Chrome is. I guess all anyone cares about is who's "winning" the browser "wars." No surprise it's Google, with their own major OS (Android), a deal with the makers of another major OS (Micro$oft Windows), and their joint control of much of the Web. Yes, Firefox offers an alternative, but there's not much advantage in running it these days. There's still Apple Safari, I suppose, but (typical of Apple) it's closed-source, isn't it? I think the last version of Safari to run on XP is as ancient and unusable as IE 8.
  13. Mathwiz
    @xpuser33, I think you may misunderstand @roytam1's role. He isn't writing browsers from scratch; he's merely bringing them to the XP/Vista community from other platforms. You may also misunderstand the role of the browsers you tried. They are not modern browsers and will fail on many modern Web sites, mostly due to changes in Javascript libraries used by many Web developers. They're based on older Firefox versions and are intended to provide a fast, "lightweight" browser for the sites that still work on those older versions. Roytam1 has some ability to fix minor bugs on his own, but it sounds like you're asking for a major redesign. If you need support for modern Web sites, I'd encourage you to try one of his UXP browsers instead: New Moon 28 or Serpent 52. (For your convenience I linked to the latest versions for non-SSE2 CPUs.) Unfortunately some Web sites won't render properly even on these newer browsers, and of course they'll run more slowly, but I think you'll have much better luck on many sites with these newer browsers. If the sites you need won't work on these newer browsers either, you'll have to try one of the many flavors of 360EE instead. Almost all sites will work correctly on 360EE v13, but it is something of a memory hog.
  14. Mathwiz
    In retrospect, I wonder if Mozilla regrets jumping on the Google bandwagon? You're not gonna out-Chrome Chrome, and they probably didn't endear themselves to their user base by dumping their entire add-on ecosystem in one fell swoop. Multiprocess mode is nice, but users deserve a choice and weren't given one. Of course, a little perspective on those numbers would be nice. Although I'm sure most of those lost users switched to one of the Chromium variants, I wonder how many moved to other FF variants, like Mozilla's own SeaMonkey, MCP's browsers, Waterfox (Classic and modern) and the like.
  15. Mathwiz
    I don't think that was what he meant. Sounded to me like he thinks the Rust language (used for FF 56+) has some inherent vulnerabilities, regardless of OS. I do run MBAE on my XP VM; hopefully that's good enough....
  16. Mathwiz
    More modern browser options for XP are certainly a good thing. And it's also good to hear that it's faster than 360EE. And it does my heart good to see @feodor2 stick it to the jerks at MCP by just forking Firefox itself But don't let this become an excuse to cease development of @roytam1's current browsers! Remember, from FF 57 forward, Firefox no longer supports the vast majority of add-ons we've all been using. Every add-on will need to be replaced with a "Web Extensions" version - if it exists. Otherwise you're SOL. An XP fork of "classic" Waterfox (before they got sold) would solve the add-ons problem, but I don't know if that's in his plans. Also, while not a big deal for most users these days, be aware that FF 91 no longer supports Adobe's Flash plug-in. If you run Flash content you'll need to keep an older browser around.
  17. Mathwiz
    So are mine, as it happens. I never got around to uploading a picture of myself in them, though. I used to have a profile pic, but it wasn't a picture of me - just a parody of the old Enron logo after that company defrauded itself into bankruptcy. At any rate, it went away some years ago, and rather than putting up a self-portrait, I just left the big "M" in its place.
  18. Mathwiz
    Yes, it is. Often it's a kind of shorthand: I often say, "this program is stupid" when what I really mean is "this program was designed stupidly." Similarly, I think describing an OS as "evil" is usually shorthand for ascribing evil motives to its designers. In the case above, though, I think "ugly" may have been a better adjective, since the complaint was about appearance, not (say) telemetry. We'll never all agree on which OS is the most pleasing to the eye, let alone "best." To each his own; that's one reason MSFN exists!
  19. Mathwiz
    That is correct. The problem appears to be that if it's too recent, it seems to trigger that blasted "[0x80072F8F] Your computer's date and time appear to be out of sync with an update certificate." The original CA.crt ran from 2015-2025 and seems to work just fine. Yours runs from 2020-2030 and also seems to work fine. Dave's runs from 2021-2031 and does not seem to work. I also had one that ran from 2021-2031 that didn't work. I had to replace it with the original 2015-2025 one to fix it. I assume a new one was created when @Dave-H installed @Thomas S.'s version, which presumably runs from 2022-2032. It doesn't work either.
  20. Mathwiz
    So, maybe 2020-2030 works too? If so that would be better than 2015-2025; it would give us another five years, assuming the remnants of WU/MU last that long....
  21. Mathwiz
    Ironically that may be the solution: CRTs just don't "do" sub-pixel font rendering. I guess in theory they could, but you'd have to set up a custom screen resolution that exactly matched the details of the CRT's shadow mask, and only a Trinitron-style CRT could possibly work with ClearType. OTOH, a CRT may make things worse on an OS that "assumes" sub-pixel font rendering is always possible. FWIW, I don't think Win 7 is "evil," but if you can't stand to look at it, it's pretty useless to you. (I do have a few very old but working CRT monitors lying around; if anyone wants one PM me. Pay for shipping and they're yours.) As for myself, sub-pixel font rendering has always looked great as long as the colors involved were black and white; but when other colors get used, it starts to look pretty awful. Looks fine. Unfortunately we're all trying to render that image on our own monitors, so it may not look to us the same way it looks to you! I had no idea the vagaries of font rendering were causing so many of us so much grief. Too bad M$ isn't listening. Now, what were we talking about again? Oh, yes.... I never thought I'd say this, but thank M$ for IE!
  22. Mathwiz
    I see D.Draker uploaded it for you. (Thanks!) It won't make any difference, though, unless you need it to validate another certificate that was signed by that one. I understand. I don't think there's any problem with ProxHTTPSProxy. I think there's a bug with Windows Update when it tries to validate a site certificate signed by your ProxHTTPSProxy certificate that runs from 2021 (last year) to 2031. It's throwing that date/time error code when it shouldn't be; your ProxHTTPSProxy certificate is fine but WU still chokes on it. As to actually finding that bug and fixing it, where's @mixit when you need him? For some reason, WU seems to like the original ProxHTTPSProxy certificate (the one that runs from 2015-2025) better. But, it won't work unless you put it in your trusted root store (with a command like the one D.Draker gave you for the M$ certificate) and also recreate your site certificates. I think ProxHTTPSProxy will re-create your site certificates automatically if it sees that CA.crt has changed, but just in case, you can rename your ...\Certs folder and create a new, empty one. That will force ProxHTTPSProxy to generate all new site certificates and sign them with the new CA.crt. The reason for renaming the folder vs. just clearing it is just for performance; if things go wrong, you could just delete everything in ...\Certs, but if instead you go back to your current configuration, ProxHTTPSProxy won't need to re-create all your site certificates yet again. If you have to put your current CA.crt back, it can just go back to using the old ones. A final note: CA.crt contains both a certificate and its private key. The private key is needed to sign the site certificates that ProxHTTPSProxy creates. From a pure security standpoint, it's unwise to share a file with a private key like CA.crt, because (in theory) any two folks using the same CA.crt could decrypt each other's communications, if they had access to each other's computers. But for what we're doing, it's probably fine. I doubt that any of us is inclined to spy on anyone else! That said, does anyone using ProxHTTPSProxy have a CA.crt expiring between 2025 and 2031? It'd be interesting to see if it works or not. I suspect there's a particular point in between (1/1/2028?) where things go wrong.
  23. Mathwiz
    Clean Flash Installer version 34.0.0.211 is now available at https://gitlab.com/cleanflash/installer/-/releases/. Confirmed working on Windows XP. Don't forget you need JustOff's GitHub-wc-polyfill add-on installed to access GitLab from a UXP browser; also Serpent 52 must be in single-process mode. 360EE should work without issue.
  24. Mathwiz
    I have Windows 7 too, which actually isn't too bad (someone once joked that Windows 7 was just Vista SP2; M$ just bumped the version so they could charge you for it again). And I do have a copy of ChrEdge on it, although I rarely use it anymore - I just use the same 360EE v13 as on XP.
  25. Mathwiz
    Well, at least Micro$oft uses Javascript to remove the "browser not supported" banner, rather than just looking at the User Agent and "assuming" the browser isn't supported! Most sites just don't work at all if the Javascript errors out, and/or give "browser not supported" if the UA is unexpected, even if the page's Javascript does work! The https://docs.microsoft.com/en-us/windows/win32/api/winuser/nf-winuser-createwindowexa page renders correctly in 360EE v13, but shows "browser not supported" in Serpent 52 even with 360EE's user agent. That said, there's nothing on that page that requires any particularly advanced Javascript. Micro$oft just wants you to use their Edge - and of course a Windows version that supports it (not to mention a PC that will run that Windows version). How dare you try to browse a Micro$oft site with XP!
×
×
  • Create New...