Notices by Wez Furlong :terminal: (wez@fosstodon.org)

@b0rk @ramsey @janl it was easier then also because it was essentially just simple HTML, maybe a dash of CSS and only if you were fancy, some JS. All those things were smaller in scope than today. On the server side it was just PHP and some kind of SQL database. All so much more grokkable than the fullest of the full stacks that folks deal with today. I think that modern webdev is dizzying in its complexity; it's very intimidating compared to what it once was. Much harder to get started today!

@b0rk (I also learned this around the same time as @ramsey and @janl) it was easy in the sense that you just edited a file and copied it up (or edited live on the box!). You didn't *have* to think about version control or push practices because the discipline was in its infancy and those things were not super common. It would have been more difficult to do those best practices because they were not common.

@rain One of the things I recognized at the start of my career in software consultancy was the difference between what someone *says* that they want and what they *actually need*. Understanding the difference and conveying that difference in a way where they don't feel like you told them "no" or otherwise in an adversarial way is its own skill and is an important part of being a lead on a technical product, not just in devtools.

@rain in devtools, one of the big problems can be summarized by the saying "doctors make the worst patients". Software engineers/programmers are the worst customer of a software product because they think they know and understand everything of relevance, when in reality they lack key domain experience/knowledge. Their influence and confidence (just a quick PR to add a feature for a power user...) can lead to those "bad" features that are then very difficult to unpick and put "right".

@shiflett I always thought this read as bum notice, which struck me as an odd title for a show

@b0rk FWIW, the cause of this is usually due to the program using a simple line reading routing (eg: `fgets`), or using a GNU-readline-compatible library like libedit that isn't correctly setup (the pgsql repl used to behave like this in the past, unsure if this is fixed by default now), or because there is some mishap with TERMINFO/TERMCAP that is breaking the line reading library.

@b0rk the EdenFS portion of Sapling (https://github.com/facebook/sapling/) has the guts of this inside. Even though that project was built primarily for Sapling (an evolution of Mercurial), it can also mount Git repos. We talked about exposing literally the interface your described for some internal infrastructure in the early days. I'm no longer part of that team so I'm not sure if that ever got implemented, but it wouldn't be difficult if someone was motivated!