Jump to content

Mathwiz

Member
 View Profile  See their activity

  • Posts

    1,737

  • Joined

  • Last visited

  • Days Won

    49

  • Donations

    0.00 USD 

  • Country

    United States

 Content Type 

Everything posted by Mathwiz

  1. Mathwiz
    That's weird; I tried that too, but the problem didn't disappear for me. Oh, well; at least others can reproduce the problem now!
  2. Mathwiz
    I used to like DST, but that was back before so many devices depended on accurate time. Now it's just a hassle, and I think most of the world should go back to standard time year round. A bill was introduced in the US Congress to put the entire nation on permanent DST, which would at least stop the semiannual changes, but I'm opposed - because it means I'll never get back the hour of sleep I lost in March! I'm owed that hour, so first, let's go back to standard time, then let's stop the semiannual changes!
  3. Mathwiz
    @jmeno, Firefox (and forks of it such as PaleMoon) have their own certificates independent of Windows.
  4. Mathwiz
    On my system they are at C:\Users\<my user name>\AppData\Local\Microsoft\Windows\Themes\
  5. Mathwiz
    I don't think @looking4awayout is around any more. His last post on this forum was Sept. 2020. But I think you can still tailor his UOC patches as you wish. They are merely text files containing very simple JavaScript: just a series of calls of the pref function to set default preferences before the user preferences in your profile are processed. You can edit them or create your own with any text editor - even Notepad.
  6. Mathwiz
    Found another possible clue. All the icons I'm having problems with post-4/29 are contained in files found at <serpent install directory>\browser\omni.ja\chrome\browser\skin\classic\browser and named Toolbar*.png. The .png appears to be a simple left-to-right array of icons, with the "oops" icon seen in the above screen shot as the left-most icon. So this seems to be some kind of indexing error, where all the icon indexes somehow get reset to zero (but only on my home PC, not my work PC)! Maybe it's as simple as my theme? BRB.... Edit: Nope, that didn't work. Actually, I should have known, because I already tried a clean profile, which of course uses the default theme.
  7. Mathwiz
    It looks like Mozilla at least keeps the user informed and in control: That's the way it should be. You can disable it, or choose a DoH provider other than the default Cloudflare. I don't entirely trust Mozilla, but I trust them more than Google!
  8. Mathwiz
    On a brief off-topic note, do you know the last Chrome version (and perhaps more importantly, which version of Android System WebView) where this "feature" either doesn't exist or can be disabled? It's important for my Android devices. (If you reply, you may do so privately if you wish.) Back on topic, this isn't a big deal for XP/Vista users, as I don't think there is (or will be) a Chromium 103 that's back-ported to those platforms. I'm not even sure if Windows 7 will run Chrome 103! And users of newer Windows versions can simply avoid the update and stay on 102 - at least until Google cooks up yet another JavaScript "improvement" that all the "major" browsers will adopt - right after they all switch to built-in Secure DNS. I don't oppose Secure DNS, but the end user needs to be in control of it - particularly which DNS servers it uses - not the likes of Google. Ideally, I'd like to implement it from my Pi-Hole to the Internet, but keep the devices on my home network using "traditional" DNS to the Pi-Hole. I would hope that "unGoogled" Chromium will retain at least the ability to disable it. But I'll be very interested in how Mozilla chooses to implement Secure DNS.
  9. Mathwiz
    They're not needed with @roytam1's Mozilla-based browsers, which have their own up-to-date TLS support built in. So the topic doesn't come up often in this thread. But both IE 8 and Chromium-based browsers use the TLS support built into Windows. The only use case for IE 8 these days is Windows/Microsoft Update and the Update Catalog, but the modern Internet often needs a Chromium-based browser on WinXP (to support the increasingly "Googlized" JavaScript language), yet also cryptographic functions that WinXP doesn't support. Although ProxHTTPSProxy and its brethren were originally designed to use Proxomitron with secure sites, they have the useful side effect of supporting cryptography beyond what WinXP has built in, so you can browse sites that require newer cryptographic functions with Chromium-based browsers. 360EE may be a special case; since it was designed to bring newer Chromium versions to the WinXP platform, it may have more up-to-date TLS support built in. I don't know. That's because you don't know its true purpose. It's been sold as a way to prevent eavesdroppers (including your ISP) from learning what sites you visit, and it does do that; but I believe its true purpose is a bit more sinister. I'm sure lots of folks here have used their "hosts" file to deliberately block telemetry, malware, or advertising servers. It's a simple but often quite effective tactic. There's even a network-wide version of this trick, called a "Pi-Hole," that's a deliberately defective DNS server to block those servers across your entire home network. These tricks have become a concern for the Googles of the world, whose business models rely on collecting our data and serving ads; so much so that some of them are building their own DNS clients into their adware to bypass any such misdirection. The built-in server always goes to the IP address of a well-known DNS server, thus bypassing any hosts file entries, Pi-Holes, or what have you. But that trick can be blocked too - at the router! The router can just be programmed to route requests to, say, 8.8.8.8, over to your local Pi-Hole instead. So the ad servers still get redirected and blocked. That's where Secure DNS comes in. If you redirect Secure DNS requests, the client will know about it, because the certificate will be incorrect. At that point, the client just says, "if I can't connect to my real DNS server, I'm not going to run at all! You either accept my telemetry and ads, or you get nothing! Good day, sir!"
  10. Mathwiz
    Make sure NM 28 isn't configured to go through ProxHTTPSProxyMII or a similar secure proxy. (NM 28 shouldn't need it.)
  11. Mathwiz
    Cleaned (i.e., adware/spyware removed) versions of the "Chinese thing" are available at https://gitlab.com/cleanflash/installer/-/releases/ Note: Gh-wc-polyfill v1.2.19 is now required for gitlab.com to work, so install that first. Should install without modification on St 52; due to version numbering differences between PM and NM, NM 28 will require modifying install.rdf in the .xpi file (which is actually just a .zip file with a different extension). Edit: I now recommend Palefill v1.20 (link is to @AstroSkipper's version) over Gh-wc-polyfill. Works with more sites and is compatible with all of @roytam1's UXP browsers without modification. (Modifying install.rdf is still required for Serpent 55, which is not UXP-based.)
  12. Mathwiz
    So, it looks like XP users will finally have another browser supporting JavaScript's new ?? operator. Now if only we could get developers to give it a plain-English (or plain-any-language, for that matter) name. Just to clarify, I was being curmudgeonly about "nullish," which isn't a, you know, word, at least not in the English language. It only exists in the context of Javascript programming. Don't believe me? Do a Google search! According to MDN, So in defining their non-word, they used another non-word: "falsy!" I'm sorryish, but this is just stupidy. Edit: to be fair, the terms "truthy" and "falsy" (the latter not to be confused with the actual word falsie) have been around for at least a decade. They seem to be used not only with JavaScript, but also other dynamically-typed languages like Python and, AIUI, Ruby. These languages have the Boolean values true and false but also values (such as numeric values) that evaluate to true or false when used in a Boolean context. (And "truthy" may have been a back-formation paying homage to Stephen Colbert, who coined the neologism "truthiness.") "Nullish," however, still just sounds silly....
  13. Mathwiz
    You know, in theory I share MCP's "ideological aversion" to DRM, but I've never understood how their refusal to support it is supposed to further their presumed goal of discouraging DRM's use! Seems to me all it does is discourage the use of Pale Moon and its forks. It's not as if PM's lack of DRM support has any content provider thinking, "I'd better not DRM my content, or I'll lose all three of my PM browser users!"
  14. Mathwiz
    I've read that JustOff is Ukrainian, and so would have bigger issues than maintaining browser add-ons to deal with these days. One hopes he is well and safe....
  15. Mathwiz
    I'm sure he'll consider it; in the meantime, have you tried IceDove?
  16. Mathwiz
    OK, have not done that. From your query, I take it the button icons are actually a customized character font, much the same as many Web sites (including MSFN) use to add small icons to their sites. Let me fire up the 32-bit version of St 55, which still shows the bug on my home PC, and see if a similar bug affects the icons at the top of the MSFN page. Hmm.... MSFN's icons appear normal, even though St 55's icons are fouled up: Still, I think you're onto something. Just can't quite figure it out yet.
  17. Mathwiz
    Well, no; at least, not that I know of. I'd know if I did that, right?
  18. Mathwiz
    This issue has me even more puzzled than ever. Here's what I've run into so far: I still see the problem in the latest St 55 version. On my home PC running Windows 7, both the 32 and 64 bit versions of St 55 exhibit the problem, regardless of profile, and even in "safe mode" (add-ons disabled). On my work PC, I tested the 64 bit version under Windows 7 and the 32 bit version under Windows XP. Both worked fine. The screen shot above was from my work PC. All versions (working and not) were tested with @looking4awayout's UOC patch, but removing it turned out to have no effect either. My home PC has text set at 150% while my work PC is at the default 100%. But setting my home PC text size to 100% had no effect either. I was able to regain the button icons on my home PC by replacing basilisk.exe, xul.dll, and a couple of .ini files, with the 4/29 versions of those files. Of course, that reverts many of the fixes @roytam1 made since 4/29, so it's definitely not a good solution. Reverting other files, without reverting those, doesn't help (I tried them all). Conclusion: something about my home PC, in conjunction with one of the many changes made between 4/29 and 5/6, causes this problem. Nothing else seems to matter. My home PC has an oldish AMD 2-core CPU with an integrated ATI Radeon GPU. My work PC is a Dell with a slightly newer Intel 4-core CPU; not sure about the GPU. BTW, I consider myself lucky that reverting files worked at all; I was just trying to narrow down the location of the problem.
  19. Mathwiz
    I second that motion. The change log is usually quite long and very technical; hard to understand even for those of us who dabble in programming. Putting it in a spoiler tag would let us refer to it if need be but save our scroll wheels if not.... Even better would be a plain-language "executive summary" of the changes, but I'm sure that's out of the question. It would have to come from upstream....
  20. Mathwiz
    Yes, that's correct; @UCyborg had an issue with 52; I had one with 55.... But oddly, I reinstalled the UOC Patch to start searching for the problem, and my button icons are still OK. So ... never mind? I'm so confused....
  21. Mathwiz
    I suspect @looking4awayout's UOC patch. The way it's installed (copy to defaults\pref subdirecory), it affects all profiles, including clean ones. I reinstalled the 2022/05/06 version without the UOC patch and the button icons are normal. Now I just have to find out which of the many, many prefs changed by the UOC patch suddenly started malfunctioning in the 2022/05/06 version....
  22. Mathwiz
    Yes it does: "Safe mode" doesn't solve the problem either.
  23. Mathwiz
    Some button icons seem messed up in the latest (2022-05-06) St 55: What it looks like in earlier versions (this is from 2022-04-29):
  24. Mathwiz
    Hey - I expect to be in my 3rd or 4th robot body by then!
  25. Mathwiz
    What is it with us programmers anyway? Not only do we have to mess around with a perfectly reasonable programming language, thus requiring constant browser updates, just to save a few keystrokes on occasion.... ... we also have to make up new words just to describe what we're doing!
×
×
  • Create New...