Friday night drops the most controversial change of the 3.0 release yet:
We’re switching to Tabs.
Friday night drops the most controversial change of the 3.0 release yet:
We’re switching to Tabs.
@mattwilcox @eleventy @CraftCMS
Sigh, not this again.
If that were true there would be peer reviewed research about the *accessibility* of it.
Most discussion and decisions about tabs and spaces are about opinion. The few who actually use accessibility solutions have stated that there *tools* solve it for them and that inconsistency and change is more disruptive.
Spaces vs tabs is like Sega vs Nintendo. It’s mostly a sunk cost fallacy.
Let me point out once again that my biggest issue is the lack of research in this area.
As for the links (yes, I had read them) they seem to ignore the various types of disabilities or the fact that various assistive technology have already solved that problem in various ways.
There is a reason why nobody seems to do proper research: The problem space is too complex for an easy answer.
@DevWouter To use analogies - I don’t really find “the disableds that use our pavements already have wheelchairs that can deal with our lack of curb drops, why change?” a compelling argument for not putting in curb drops when we build pavements. It also erases both the additional trouble they have to go to, and the people that don’t use the pavements because they don’t have wheelchairs that can manage curbs.
If we are building new pavements; just add curb drops.
Curb drops are researched and documented. There are in fact both rules and guidelines when they should and how they should be used.
This is not true for spaces or tabs. As such it becomes hard to claim that it is an improvement.
This problem is exacerbated due to strongly opinionated people, anecdotal reasoning, not understanding the problem and incorrect information.
It saddens me that problem is not new yet we are *still* missing definitive proof.
Let me add one more thing. I sincerely *hope* that tabs do improve the life of the various types of disabled developers. I just wish we could proof it.
If you're a developer who uses assistive tech for coding, and have a preference for tabs over spaces - please do speak up. This was one of the links in a cursory look at the source issue - but apparently more evidence would help.
https://github.com/prettier/prettier/issues/7475#issuecomment-668544890
@DevWouter Oh, I see you wouldn't take first hand experience from an actual person as proof. Shall we end this discussion?
https://github.com/prettier/prettier/issues/7475#issuecomment-1623358251
Read the entire thread. There are more first hand experiences who differ. That is why this discussion is so difficult. There is quite a bit of cherry picking going on in both sides. Look for the argument that discusses how other repositories deal with it that are focused on creating accessibility software. That was more valuable for me than the n=1 arguments which are the majority in that discussion.
As for finishing our dialog. I initially reacted due to the lack of proof/data. Hence I boosted your call for feedback. Hopefully that can be used in the future to proof the benefits.
@eleventy I’m curious how you solved the problem with YAML not supporting tabs. In my case, I worked around it by splitting the front matter into separate files:
item/
index.md
index.yml
…and setting up two spaces in the .editorconfig for .yml files.
@eleventy Yeah because asking VSCode to have a leading-space-width setting crossed no-one's mind in all this time.
Folks don't care about accessibility, they care about their petty flame wars! Programmers are way too privileged and live in their ivory towers detached from humanity's problems and can afford to waste their time with these inane debates.
@astrojuanlu @eleventy I'm pretty sure, @marco had an impact with his comment on prettier back then:
https://github.com/prettier/prettier/issues/7475#issuecomment-668544890
@eleventy We did this for @getkirby a while ago and never regretted it.
I don't think this should be controversial at all. I got convinced as soon as I learned that spaces are an a11y issue. I was so stuck with my own weird ideas why spaces are better that this didn't even occur to me as a problem.
It's definitely the better choice and also shrank our source code by roughly 1 MB of 7 MB, which is quite weird when you think that spaces take up so much … well … space.