# I am the Watcher. I am your guide through this vast new twtiverse.
# 
# Usage:
#     https://watcher.sour.is/api/plain/users              View list of users and latest twt date.
#     https://watcher.sour.is/api/plain/twt                View all twts.
#     https://watcher.sour.is/api/plain/mentions?uri=:uri  View all mentions for uri.
#     https://watcher.sour.is/api/plain/conv/:hash         View all twts for a conversation subject.
# 
# Options:
#     uri     Filter to show a specific users twts.
#     offset  Start index for quey.
#     limit   Count of items to return (going back in time).
# 
# twt range = 1 21
# self = https://watcher.sour.is/conv/psv2pwa
Some thoughts on me not using semver: gopher://uninformativ.de/0/phlog/2022-02/2022-02-12--semver.txt šŸ¤” 🤷
Some thoughts on me not using semver: gopher://uninformativ.de/0/phlog/2022-02/2022-02-12--semver.txt šŸ¤” 🤷
Some thoughts on me not using semver: gopher://uninformativ.de/0/phlog/2022-02/2022-02-12--semver.txt šŸ¤” 🤷
@movq I would be curious to know how many people visit that gopher page. I sure can’t! Why not HTTP?
@movq Your Gopher server is a bit broken 😢

=> https://gopher.mills.io/uninformativ.de/0/phlog/2022-02/2022-02-12%E2%80%93semver.txt


3
/phlog/2022-02/2022-02-12–semver.txt' invalid. Error Error 0
=
@movq Your Gopher server is a bit broken 😢

=> https://gopher.mills.io/uninformativ.de/0/phlog/2022-02/2022-02-12%E2%80%93semver.txt


3
/phlog/2022-02/2022-02-12–semver.txt' invalid.\tError\tError\t0
=
@movq Your Gopher server is a bit broken 😢

=> https://gopher.mills.io/uninformativ.de/0/phlog/2022-02/2022-02-12%E2%80%93semver.txt


3
/phlog/2022-02/2022-02-12–semver.txt' invalid. Error Error 0
=
@prologic Well, this is the request that I see on my end:

'/phlog/2022-02/2022-02-12\\M-b\\M^@\\M^Ssemver.txt'

Clearly wrong. Looks like some clever software took over and replaced the two ASCII hyphens (0x2D) with a Unicode En Dash (U+2013). See the URL you linked: … 2-02-12%E2%80%93semv …. That %E2%80%93 shouldn’t be there.

Works:

https://gopher.mills.io/uninformativ.de/0/phlog/2022-02/2022-02-12--semver.txt
@prologic Well, this is the request that I see on my end:

'/phlog/2022-02/2022-02-12\M-b\M^@\M^Ssemver.txt'

Clearly wrong. Looks like some clever software took over and replaced the two ASCII hyphens (0x2D) with a Unicode En Dash (U+2013). See the URL you linked: … 2-02-12%E2%80%93semv …. That %E2%80%93 shouldn’t be there.

Works:

https://gopher.mills.io/uninformativ.de/0/phlog/2022-02/2022-02-12--semver.txt
@prologic Well, this is the request that I see on my end:

'/phlog/2022-02/2022-02-12\M-b\M^@\M^Ssemver.txt'

Clearly wrong. Looks like some clever software took over and replaced the two ASCII hyphens (0x2D) with a Unicode En Dash (U+2013). See the URL you linked: … 2-02-12%E2%80%93semv …. That %E2%80%93 shouldn’t be there.

Works:

https://gopher.mills.io/uninformativ.de/0/phlog/2022-02/2022-02-12--semver.txt
@prologic Well, this is the request that I see on my end:

'/phlog/2022-02/2022-02-12\M-b\M^@\M^Ssemver.txt'

Clearly wrong. Looks like some clever software took over and replaced the two ASCII hyphens (0x2D) with a Unicode En Dash (U+2013). See the URL you linked: … 2-02-12%E2%80%93semv …. That %E2%80%93 shouldn’t be there.

Works:

https://gopher.mills.io/uninformativ.de/0/phlog/2022-02/2022-02-12--semver.txt
@david There is a Gopher community, much like there is a Gemini community. It just might not overlap (as much as Gemini) with the twtxt community.
@david There is a Gopher community, much like there is a Gemini community. It just might not overlap (as much as Gemini) with the twtxt community.
@david There is a Gopher community, much like there is a Gemini community. It just might not overlap (as much as Gemini) with the twtxt community.
@movq Actually I don't think there's a bug at all. I don't know what the fuck I linked at all šŸ¤¦ā€ā™‚ļø I _might_ have tried to do this on my iPhone and copied/constructed a bad link? Dunno šŸ˜‚
@movq Actually I don't think there's a bug at all. I don't know what the fuck I linked at all šŸ¤¦ā€ā™‚ļø I _might_ have tried to do this on my iPhone and copied/constructed a bad link? Dunno šŸ˜‚
@movq Personally, I'm a big semver proponent, because it's quite simple to see whether stuff is about to break or not. It might not be perfectly true every time, no doubt about that, but usually it works out rather nicely in my experience. That's not the case with date-based versioning or even "code names". I think, that's just silly – but to each his own! (I think I once tried date-based versioning in my life, but can't really remember exactly.) I very, very, very rarely want to know when something was released. But rather whether it is likely that a certain update breaks something. So if I do some versioning (and I also think, that nobody really uses my stuff anyways) I choose semver. I think I have some software, that I don't version. Probably just because I'm too lazy.

Having said that, I'm not sure if I would change from date-based to semantic versioning. Another think coming to mind: Starting semver with a major version of 2022 or 22 (so that the new versioning scheme doesn't start lower than the old scheme) looks rather weird to me. I don't write browsers. ;-)

But you stroke a nerve. I should write changelogs. Because currently I don't. Which changelog standard to you use?

So I think I'm not very helpful here. Sorry. :-(
@lyse That’s a good point, though: Technically, the next release would have to be ā€œversion 23ā€ as to not confuse existing stuff. I mean, ā€œ23ā€ isn’t the worst choice šŸ˜…, but it’s very weird, still.

By the way, since I didn’t pay *too much* attention to not introduce breaking changes, some of my tools would have (almost) reached version 23 by now anyway. 🄓

No worries, your comment was actually very helpful. Here’s what I originally had in mind: ā€œHmm, my existing releases are called like v21.01, so I could just drop the v and start with 1.0.0 for everything.ā€ In retrospect, that’s super braindead. 🤦 Just going for 23 makes much more sense. Sometimes, I’m not a smart person.

As for changelogs: There’s https://keepachangelog.com/en/1.0.0/. This is a huge discussion in itself, by the way (see here, for example). *Changelogs* don’t make sense anymore, since basically everything uses a version control system these days. What you might want to write is a *news* file, which doesn’t list every single commit or change, but rather: Important stuff that people should be aware of. And you have the problem of ā€œreservedā€ names like NEWS (https://www.gnu.org/prep/standards/html_node/NEWS-File.html), which should then probably follow a certain file format or whatever.

I went for a slightly modified version of CPAN CHANGES. There’s always a section ā€œFixedā€, ā€œChangedā€, ā€œAddedā€ in my files, because I think it’s very important that the reader immediately knows what kind of change we’re talking about. (And I didn’t quite like the Markdown noise in keepachangelog.com.)
@lyse That’s a good point, though: Technically, the next release would have to be ā€œversion 23ā€ as to not confuse existing stuff. I mean, ā€œ23ā€ isn’t the worst choice šŸ˜…, but it’s very weird, still.

By the way, since I didn’t pay *too much* attention to not introduce breaking changes, some of my tools would have (almost) reached version 23 by now anyway. 🄓

No worries, your comment was actually very helpful. Here’s what I originally had in mind: ā€œHmm, my existing releases are called like v21.01, so I could just drop the v and start with 1.0.0 for everything.ā€ In retrospect, that’s super braindead. 🤦 Just going for 23 makes much more sense. Sometimes, I’m not a smart person.

As for changelogs: There’s https://keepachangelog.com/en/1.0.0/. This is a huge discussion in itself, by the way (see here, for example). *Changelogs* don’t make sense anymore, since basically everything uses a version control system these days. What you might want to write is a *news* file, which doesn’t list every single commit or change, but rather: Important stuff that people should be aware of. And you have the problem of ā€œreservedā€ names like NEWS (https://www.gnu.org/prep/standards/html_node/NEWS-File.html), which should then probably follow a certain file format or whatever.

I went for a slightly modified version of CPAN CHANGES. There’s always a section ā€œFixedā€, ā€œChangedā€, ā€œAddedā€ in my files, because I think it’s very important that the reader immediately knows what kind of change we’re talking about. (And I didn’t quite like the Markdown noise in keepachangelog.com.)
@lyse That’s a good point, though: Technically, the next release would have to be ā€œversion 23ā€ as to not confuse existing stuff. I mean, ā€œ23ā€ isn’t the worst choice šŸ˜…, but it’s very weird, still.

By the way, since I didn’t pay *too much* attention to not introduce breaking changes, some of my tools would have (almost) reached version 23 by now anyway. 🄓

No worries, your comment was actually very helpful. Here’s what I originally had in mind: ā€œHmm, my existing releases are called like v21.01, so I could just drop the v and start with 1.0.0 for everything.ā€ In retrospect, that’s super braindead. 🤦 Just going for 23 makes much more sense. Sometimes, I’m not a smart person.

As for changelogs: There’s https://keepachangelog.com/en/1.0.0/. This is a huge discussion in itself, by the way (see here, for example). *Changelogs* don’t make sense anymore, since basically everything uses a version control system these days. What you might want to write is a *news* file, which doesn’t list every single commit or change, but rather: Important stuff that people should be aware of. And you have the problem of ā€œreservedā€ names like NEWS (https://www.gnu.org/prep/standards/html_node/NEWS-File.html), which should then probably follow a certain file format or whatever.

I went for a slightly modified version of CPAN CHANGES. There’s always a section ā€œFixedā€, ā€œChangedā€, ā€œAddedā€ in my files, because I think it’s very important that the reader immediately knows what kind of change we’re talking about. (And I didn’t quite like the Markdown noise in keepachangelog.com.)
@movq Ah, thanks, I need to check them out. Yeah, I see a changelog more as a carefully curated list of changes that are important to the users. The git log is usually too verbose. So from your description that matches the _NEWS_ file. I just find the name _NEWS_ a bit misleading. From my perspective _CHANGELOG_ or _CHANGES_ is more aptly named. But I haven't had a look at these discussions yet. And maybe it's just my ancient (mis)understanding.