I definitely try to avoid any public statement of political nature online. You never know how the tide will turn at some point and who gets into power. And then you do not want to have a record of having said the wrong thing about the new guy(s) at the top in your past.
What is that? Like 32x >100MB junk overhead per app? ~4GiB gone from the disk just to hold the same broken copy of some framework/library? It's quite the insanity, isn't it?
If there was one copy of that electron (e.g. installed to /Library somewhere) which all apps would simply use then you only would need to update one copy. Less disk space wasted. All apps fixed in one go.
Back in the old days on the Commodore Amiga we would just do that… install some .library to SYS:Libs/ first if a program required it. It's not like this process was so complicated nobody could do it, right?
There is not one Electron. There are multiple, they release a new version every month or so.
Some apps, like VS Code, update very quickly to the latest one. Others more rarely. So now you need to keep multiple shared Electron versions, and track dependencies and who uses what version.
And it's quite likely that everyone of your Electron using apps will be on a different version, so now you are back to square one.
Memory and storage is cheap enough nowadays to not have to deal with the insanity that shared libraries cause. I don’t care if I use 30gb of memory to run a browser and a note taking app.
Some of us do care. Devs should respect users systems more than their own instead of crapping all over them with Electron. I’d almost go as far as to say that it’s evil. Wasting resources, energy, people’s time, and money.
I don't understand why it's all-or-nothing. We know how to version things pretty well these days, why is there no blended solution where libraries are shared but version aware? I don't mind having two different versions of electron on my laptop but I don't want 30 copies of the same version.
The big issue I see with Nix is that it's solving several related & very complex problems, and isn't doing so at a particularly easy level of abstraction. It's a PITA to package software that isn't using an already-supported build system. And mixing versions is messy, instead of just `[ package1="versionA", package2="versionB", …]` sort of thing with a lockfile to convert versions to immutable IDs like commit hashes you have to specify which commits of nixpkgs had each version and then do something like `nixpkgs-versionA=GIT_COMMIT_HASH_A; nixpkgs-versionB=GIT_COMMIT_HASH_B; [ nixpkgs-versionA.package1, nixpkgs-versionB.package2, …]`. There are lots of other "warts" like that, of varying levels of annoyance.
Because in practice nobody has solved it, while everyone claims they have.
In practice every software needs a particular version of a library. Even a minor upgrade to that library might, and will break it. In an idealized world it should not happen, but here we are. In a world that we setup whole containers so that we can run software.
So no. Shared libraries do not work in practice. At all. It should be straightforward, but they just do not work.
No shit. One major frigging selling point of Electron vs OS web view is the developer controls the browser version and has a stable target to develop and test against, rather than having many moving targets that shift after the app is shipped.
And you really think the entire ecosystem has never heard of this honking great idea named shared libraries from the good old days? Being smug about obvious things like this usually just betrays your shallow understanding.
Disclosure: I’ve criticized Electron aplenty. But these are complex tradeoffs and people dismissing them with obvious wins in their minds are clueless.
Disclosure 2: I was once a member of the maintainer team of a major package manager. Shared libraries, oh my, I can tell you horror stories all night long. If your experiences with them are great chances are a team behind the scenes has blocked and/or fixed faulty updates for you and shielded you from most of the issues.
I read that announcement and I have zero idea what this "immich" app (?) is. How about adding some short introduction text explaining what the thing is for folks just discovering this thing via a HN link (w/o forcing them to click further or digging the information from somewhere else)?
That post complaining about lack of context took longer than typping "immich" into duck duck go and reading the single sentence that pops up at the top describing exactly what immich is.
-----------
Immich
Immich is a project that lets you back up, organize, and manage your photos and videos on your own server. It is under active development and available under GNU AGPL v3 license.
I agree that this text in its current style is very hard to read. Feels like the text was ballooned up to 3 or 4 times its original length with pointless "side content"? Lots of distracting noise basically. AI or not AI, this is not very good.
… and so I'll continue to stick with AVC, thanks! :-)
I check by sampling a piece. If you chew on it and it sticks to your teeth then it's not done yet and can cook a bit longer. No need to make a science out of it. :)
It should mention the bug only exists after some arbitrary "patch" was introduced. As the current title makes it sounds like the actual zlib has a security issue.
Seems like it's not just arbitrary, but crafted. Could not find it anywhere, for example, searching for "DISTS so we can remove overflow checks from" (with quotes ofc) brings up just this site, both in Google and Bing. It has typos, btw.
It would be another issue if it came from https://chromium.googlesource.com/chromium/src/+/HEAD/third_..., but that's not the case.
The original title included "[CTF] Google CTF 2025" which would strongly hint(CTF=capture the flag) at the possibility of an artificial setting. That probably should of been included in the submission.
Not the author. The first sentence of the article does say this “webz is a zlib exploitation challenge from Google CTF 2025. The Google-zlib implementation provided in the challenge is not upstream; it’s a version with an arbitrary patch applied.”
It’s almost quite literally your comment word for word.
I noticed it yesterday and I have to say I hate it. The "Peoples" section seems to be entirely gone (it's now intermingled with rooms in the "Home" section). You no longer can see favorite/ normal/ low priority DMs or rooms in proper order at once. E.g. in Spaces I had my one or two favourite rooms at the top in a separate section; now it's somewhere messed up in the middle with the normal priority stuff. You constantly have to fiddle with the space-wasting bubble buttons to find stuff.
No idea how that is supposed to improve things really. Why is it always necessary to break user's muscle memory with this types if extreme "improvements"?
I'm still trying to find out where to manage my Sessions in the newer Element versions (I have to keep a very old one around to do that).
I'll just go and say I hate the design changes in desktop, it's more cluttered without good reason. Feels like they made the text bolder and added a bunch of borders or increased them where they weren't before. You can't collapse the names down as far as you could in the past to save space so when this app runs tiled it's stuck with like a third less functional space than it used to have.
I'm probably going to switch back to using nheko which also has the benefit of being much lighter on resources. Nheko I can collapse the name list right down to icons if I want.
Honestly, this was the reaction I expected to have when I first started using the new roomlist (which was built by the Element design/product/eng teams, rather than the old one, which was originally written by me).
However, in practice, I've rapidly found that I genuinely don't mind the new layout, and actually prefer it. (And I tend to be have a bad attitude to change in general). The reasons are:
* It's way faster to switch rooms
* You can filter on People by hitting the People filter button rather than having a dedicated section.
* I spent my life scrolling around the previous sections, or manually resizing them, etc, whereas the simplicity actually feels a bit refreshing and lighter weight.
That said, this is an initial release of the new room list, and there's a bunch of work to be done still:
* Need to support collapsing the LeftPanel like we used to be able to.
* There's a plan to re-introduce Sections again (so you can see your favourites etc separately to the main contents if you want - and reclaim muscle memory)
In terms of where to manage Sessions in Element Web/Desktop - it should always be the 'Sessions' tab in Settings? If your server is using native OIDC (i.e. matrix-authentication-service) then some of the management gets delegated to MAS and links through to account.matrix.org (or wherever your MAS lives). But if you can't see any way to manage Sessions, then it's likely a misconfiguration or a bug; please let us know the details.
Yes, it's pretty terrible. Especially since the filters reset every time you switch spaces, so you need to set them again constantly. And filters only is just worse than dedicated sections where you can actually see multiple at a glance in smaller spaces.
This might be the point where I give up on Element (or Matrix entirely), I was willing to excuse some of the jank in exchange for it having all the features in a somewhat reasonable UI, but if they break that what's the point. Messing with it while Element X is still incomplete feels like a very weird choice.
Tried this: "yt-dlp -f 'bestvideo*[ext=mp4]+bestaudio[ext=m4a]/best[ext=mp4]/best' -S vcodec:h264 -other_options …" ? I'm still getting proper h264 with that (my Raspberry PI 3 only wants a proper codec too… none of that mushy new-era codec stuff. ;) )
Can't get past level 2 with Firefox 78 ESR. "Select all the squares with a
Stop Sign", but no squares for selecting show up, so there's nothing to select. Unless I miss a joke somewhere in there?
It took me a bit to figure out that squares stop subdividing at some point (I first thought I had to subdivide into biggest-possible squares, but no selection possible that way) — once at smallest subdivision, just select all the squares covering the stop sign.
I did hit an issue in FF on level 17 (draw a circle): mouse-down triggers a "too closely resembles a dot" and I can't draw a circle on a laptop.
Edit, with cURL it's OK… 200.
That makes no sense.