Vincent @ Vincent @feddit.nl

Posts

119

Comments

635

Joined

2 yr. ago

A comprehensive answer is out of scope and probably best given by a true accessibility specialist, but for example, if you only use <div> tags for everything, a lot of the screen reader's affordances for navigating are unusable. Images that carry information but not in their alt text are another simple example.

And then there are parts where JS could actively help. For example, if you have a tabbed interface, but clicking a tab results in a full page refresh, the screen reader loses all context.

Also keep in mind that there's more to assistive technology than just screen readers, e.g. sufficient colour contrast and keyboard navigability are important to many people. Too many websites still get those basics wrong.

Not necessarily, unfortunately. (Though I guess technically it's easier to throw up barriers using JS, but it's not an inherent quality, and leaving it out doesn't automatically make it good.)

Heh, just deleted my reply - thanks for covering all that, you're exactly right :)

They can overlap, yes. Static sites are definitely not automatically better for accessibility.

Note there's a group of users that larger than the group of users without JS (for whatever reason): users of assistive technology. And they don't even have a choice.

While I'm all for considering the needs of every user... If you get to the point where you're worrying about no-JS users, I hope you've already considered the needs of people with disabilities, whether temporary or permanent.

Edit: oh right, wanted to add: just making a site work without JS doesn't automatically make it accessible to people with special needs.

That's the kind of thing that sounds nice, but in practice I don't think that's what evidence points towards.

De partijencombinatie zegt dat het politieke klimaat is veranderd en vreest dat een minister zo'n bevoegdheid met politieke motieven kan misbruiken.

Ik ben het eens met de vrees, maar... De mogelijkheid dat het politieke klimaat verandert is er altijd al. Liefst heb ik dat mijn politici daar meteen al rekening mee houden, ipv dat we geluk moeten hebben dat het net op tijd verandert om tot dat inzicht te leiden. (Zeker omdat we er nu van afhankelijk zijn dat dat "veranderde politieke klimaat" zelf de wet moet intrekken.)

Om een motto van Bluesky te lenen: de overheid moet zichzelf als toekomstige tegenstander zien.

That is true, but all that wouldn't be able to survive if Mozilla were to significantly scale back development.

I'm not sure which button you're talking about, but if it's the one in the sidebar, click "Customise sidebar", and then uncheck "AI chatbot".

By now you would've expected someone to have pointed out what code is actually collecting that data that's supposedly sold.

I don't see how being a non-profit suddenly makes it cheaper to build a secure, modern and compatible browser. (Although I know lots of people underestimate how much effort that takes. But just consider that already Mozilla's doing it for far less money than Google invests in Chrome, for example.)

I fail to understand how they haven't figured out a way out of this seems to me they're using all the money they from Google as if there is no tomorrow... Why on earth also if Firefox is so clearly their main product does it seem to not always be at the centre of their attention?

The answer to the second question is the answer to the first - there have been a ton of attempts at alternative sources of funding, but it's hard to come close to the ~half a billion USD the default search deal provides. So far the branded services you're calling for don't seem to have been able to pull it off, and I haven't seen any signs that donations would be able to either.

(Although as for email with Thunderbird...)

I think you're grossly overestimating the share of volunteer contributions if you think it might even be over half. It's amazing what contributors do, but the vast majority, and especially thankless-but-important work like web compatibility or security, is done by paid staff.

They're saying they want something like Synaptic (mostly for its "multi-select", apparently, though I'm not sure what that means?), but have it support AppImages, Flatpaks, Snaps, etc., instead of just Debs like Synaptic does.

I only really hear 3 (or at least, hear people imply it) in reality, and maybe a small nod to 7 recently. For 4, I see people think arbitrary numbers like 90% or 80% code coverage are things that matter.

1100?! Oef, wat een nekslag!

En misschien doe ik dat wel, op zich een goed idee. De meter staat wel op een behoorlijk onhandige plaats hier, maar op zich voor een keertje is het wel de moeite waard.

Hmm bizar, heb net toch maar even mijn jaarrekening erbij gepakt, en heb zo'n €350 aan vaste kosten (dat is vastrecht, neem ik aan), en zo'n €650 aan verbruik. Ik kan me dat haast niet voorstellen.

Hmm, of I guess dat de boiler die we hebben ook de stadswarmte gebruikt om op te warmen? Is me altijd onduidelijk of dat nou puur op elektriciteit is of niet. Maar dan moet ik dus vooral korten gaan douchen I suppose.

Elektriciteit en water kosten bij mij sowieso vrijwel niks in vergelijking met warmte. Ik snap idd niet echt waar dat bedrag vandaan komt; heb 'm hier ook vrijwel niet aan gehad, was toch nog een best bedrag.

Een belangrijke reden waarom hij niet zo vaak aan hoeft is wel dat er van de buren ook veel warmte af komt, wat volgens mij bij blokverwarming wordt meegenomen, maar voor mij geldt dat volgens mij niet. (Ik heb wel stadsverwarming.) Maar volgens mij maakt het voor de kosten dus echt nauwelijks uit hoeveel ik verbruik, maar is het basisbedrag gewoon erg hoog?

Can I just say: hats off to the bug archaeology you've done there :)