James Scholes / is on Nostr

Digital #Accessibility Engineer/Analyst, #ScreenReader user, and occasional #software developer. #a11y

Last Notes

2024-09-29 00:23:52

- reply

"My own mother was confused and thought WP Engine was an official thing."
I'm now intensely curious how this conversation came about. Did it happen over a family Sunday lunch, or...?
@npub1he7…mmdk

2024-09-26 15:50:23

- reply

As a British person, this proposal initially brought some unexpected images to mind.
https://en.wikipedia.org/wiki/Co-op_Food

2024-09-06 17:23:57

Fresh out of the kitchen for my birthday: lemon and raspberry cupcakes, and mini Bakewell tarts, all with homemade raspberry jam. Who needs a traditional cake?

2024-05-26 21:32:51

- reply

In the end, it took me over an hour to extract one link from my #Mastodon favourites, as someone who already had a Python environment set up ready to run an archival tool without much fuss.

2024-05-26 20:08:50

- reply

To those who've pointed out that I should've probably been using bookmarks all along: those can indeed be exported from the account preferences section on my instance. Unfortunately:
1. I haven't been using them all along; and
2. the CSV only includes post URLs, no text or metadata, so still can't be searched, and hence doesn't come close to solving the problem of searching them.

2024-05-26 20:02:25

- reply

Don't get me wrong, you raise a good point: the archive from the account prefs includes all my posts if I want those, which makes using a third-party tool and the corresponding toll on the server doubly disappointing.

2024-05-26 20:00:31

- reply

But that doesn't include favourites... or at least doesn't say it does.

2024-05-26 19:59:29

- reply

Some replies have suggested using "in:library" as a search qualifier. Unfortunately, no matter what I specify after it, my instance claims there are no matches.
So, I'm going ahead and using the aforementioned archival tool. I don't like that it's hitting the API to obtain a lot of data that I don't want or need, but I think I've legitimately tried hard enough (short of modifying the software itself) to avoid it.

2024-05-26 19:45:54

- reply

There's also an unexpectedly large number of tools to export #Mastodon data to #Prometheus. I'm not sure I even understand what Prometheus is, nor why so many people apparently want to export Mastodon data to it.

2024-05-26 19:44:40

- reply

I found a #Mastodon archiver (https://github.com/kensanata/mastodon-archive). But while it has a `--no-favourites` flag, it doesn't have the reverse (i.e. skip my own posts and only download my favourites). It doesn't seem reasonable, in terms of instance server load, to grab a bunch of data just to throw most of it away.

2024-05-26 19:36:02

- reply

For people following along at home, I have two replies so far telling me I'm using #Mastodon wrong. In 10 minutes on a Sunday. Impressive.

2024-05-26 19:24:58

How do people meaningfully search and/or export their #Mastodon favourites? I've been favouriting (not bookmarking) a bunch of useful and/or interesting stuff since I came here, but the data seems difficult to access in a useful way.
When I use my instance's website, there's no search facility on the Favourites page. It only loads a subset of favourites upfront, so without pressing the "load more" button hundreds of times, I can't use my screen reader's find feature either.
When I try the favourites timeline of Semaphore, it also only loads a subset of my favourites, but apparently without a means of loading more of them.
In Mona on iOS, I would likewise have to keep finding, and then pressing, "load more posts".
When I try to find information on the internet about possible qualifiers for searching my own favourites, I get results about API calls and installing Elasticsearch as an instance owner.
In the "import/export" section of my account preferences, it doesn't say anything about favourites, and doesn't make it clear whether my posts archive would include them.

2024-04-09 19:34:58

- reply

I don't get it. If it replaces twitter dot com with x dot com, how does "netflix" become "netflitwitter"? For that to happen, they'd have to be replacing x dot com with twitter dot com. [@BorrisInABox](https://mindly.social/@BorrisInABox)