Notices by Jon Udell (judell@social.coop)

"I wish we had made the change earlier than we did, in a moment further from the election and the emotions around it. That was inadequate planning, and not some intentional strategy."

- Bezos editorial

"We"? You can't deflect this, Jeff, it was you.

I would love to see Taylor Swift show up at the rally today as a profile in courage standing in contrast to Bezos' profile in cowardice.

#wapo #courage #cowardice

"No major news organization has yet run even one serious story about the state of Trump’s cognition and thus of his fitness for office. We saw dozens of these about Biden; so far, none about Trump."

I hugely respect Fallows and would love to know his take on why media give Drumpf a pass, and what exactly is the turning point he discerns.

But it's a paywalled article. Can anyone who has read it summarize the arguments?

"This is yet another example of how LLMs don’t replace human developers: they augment us."

I so wish Doug Engelbart could be here too see this happening.

https://simonwillison.net/2024/Aug/8/django-http-debug/

#llm #augmentation

I try not to judge people who feel they must remain on Twitter for professional reasons.

But it's getting harder and harder to respect that choice. After a long absence I took a peek at my feed there and this is what I saw.

I'm not naming names. And there are so many of you. How, at this point, do you justify being there?

#Twitter #CognitiveDissonance #SelfRespect

"Processing unstructured data is one of the most directly useful applications of Large Language Models."

I worked hard to syndicate event data from iCalendar feeds, but they were and still are scarce. Much real-world data lives inaccessibly in event posters; that's always felt wrong.

It also feels wrong (to techies) to make feeds from pictures of posters.

"That isn't data!"

But for normies it makes total sense.

"Of course it's data, I can see it plainly!" https://fedi.simonwillison.net/@simon/112243754613807997

"You’re almost always inheriting a code base that’s been around for a long time, so the hard part isn’t writing the code to implement the feature or fix the bug. It's trying to understand why the application works the way it does, who the best person is to talk to, when these changes were introduced, and contextualizing all that to move forward."

Dennis Pilarinos on what Jack Ozzie once called context assembly, less vividly known as enterprise knowledge management.

https://www.itshipped.fm/episodes/20

@ntnsndr @hypothes_is If it becomes easier to acquire an API key for LMS-based Hypothesis, there's a lot that can be done using the Steampipe plugin and dashboards. I think relationship graphs (https://blog.jonudell.net/2023/01/16/mastodon-relationship-graphs/) could be especially helpful.

@ntnsndr @hypothes_is Hey @nathan, I think that the Steampipe plugin I wrote (https://hub.steampipe.io/plugins/turbot/hypothesis) along with a set of dashboards (https://github.com/judell/hypothesis-dashboards) would enable the kind of analytics that we were doing previously.

Per https://github.com/hypothesis/lms/issues/420, though, it looks like there still isn't a way to acquire the needed API token for a Canvas-based instance of Hypothesis. There is a hacky way to acquire the token that Michael DiRoberts could show you.

The word "content" is like fingernails on chalkboard to me, it makes me flinch. I always wish for alternative words like "news article" or "blog post" or "video" or "podcast". Unrealistic? Perhaps, I thought.

Then I asked Claude for 5 sentences that use the word and 5 rewrites that avoid it. Not so hard!

(Forgot to tell it to also avoid "creator".)

#You_Dont_Have_To_Use_The_Word_Content

"Occasionally, it’ll spit out quite a detailed solution to a coding problem I have that clearly works because I can run the code. But I won’t commit that code until I’ve at least broken it down and made sure that I fully understand it and could explain it to somebody else."

https://simonwillison.net/2023/Aug/27/wordcamp-llms/#how-to-use-them

An excellent corollary to "Never trust, always verify" (https://thenewstack.io/test-driven-development-with-llms-never-trust-always-verify/)

#llm #software #ethics

In https://thenewstack.io/should-llms-write-marketing-copy/ I recommend some ways to recruit LLMs to channel George Orwell and Strunk and White to review your writing according to similar principles.

#LLM #Writing #Editing https://zirk.us/@grammargirl/110899190161676017

@jgordon The best part of working for InfoWorld was IDG Travel. I don't think IQ is what enables us to slog through the nighmares of self-service. It's grit. And mine is wearing thin.

@bitsplusatoms @eschaton @bitsavers Agree. LLMs promise to democratize many kinds of scripting.

Here's what I've learned so far about porting list members from one Mastodon server to another.

https://www.infoworld.com/article/3692919/migrating-mastodon-lists.html