Notices by Bob Wyman (bobwyman@mastodon.social), page 2

The average cost to process a debit-card transaction in the USA is 3.9¢, yet banks have been allowed to charge up to 21¢ per transaction (over 5 times their costs). Now the Federal Reserve proposes lowering the cap on charges to 14.4¢ -- still almost 4 times their actual costs. This doesn't make sense.

Banks should be allowed to charge actual costs plus a reasonable profit. 269% is NOT a reasonable profit.

https://www.economicliberties.us/press-release/the-feds-proposal-to-lower-debit-card-swipe-fees-isnt-enough/

Pew Research shows that today a majority of Americans say a woman president would be no different than a man in several key areas.

This is a significant change from earlier years and should be celebrated.

https://www.pewresearch.org/social-trends/2023/09/27/views-of-having-a-woman-president/

@fediforum
Today's FediForum was good fun and very informative. It is great to see that so many people seem to have a sincere interest in improving the way the Fediverse works!

Question for Geeks:
XPath and JSONPath are similar, but different. That presents a problem for me since I'm building a system to query ActivityStreams objects that can include HTML wrapped in JSON.

Should I:
- Use XPath syntax for both JSON and HTML?
- Use JSONPath syntax for both JSON and HTML?
- Use either JSONPath and XPath depending on underlying datatype?

If you were writing a query, would you accept needing to know both syntaxes?

@evan @J12t
See: JSONpath: https://goessner.net/articles/JsonPath/

@J12t @evan Consider this object:
{"object": {
"type": "Note",
"content": "<p>Some Text<div>1st div</div><div>2nd div</div></p>"
}}

If I treat HTML like JSON, I can fetch the content of the first div with JSONPath like:
"$..object[?(@.type == 'Note')].content.p.div[0]"

but, the XPath for that first div would be "//p/div[1]" (Note: 1-base, not 0-base) So, should the full query be the following?:
"$..object[?(@.type == 'Note')].content//p/div[1]"

@tk @evan
I know that there was once a time when you could buy PB&J, pre-mixed in jars. The result was less than ideal... Are such pre-mixed products still available?

@jasnell @evan
Given that "Note" and it's URI are equivalent, should I read the spec as saying that I MUST use "Note," or am I permitted to designate an object as either "Note" or "https://www.w3.org/ns/activitystreams#Note"?

It might make more sense to ask: Given that all the document's examples are in Compact form, am I allowed to use other forms, such as Expanded, etc.?

@evan @jasnell
Why does the Activity Vocabulary define several specific non-URI object types instead of using URIs (e.g. media-types) to indicate types?

Given that the spec identifies a URI for each of the specific object types it defines, why does it say that I should write:
{... "type": "Note"...} instead of
{... "type": "https://www.w3.org/ns/activitystreams#Note" ...}?
Is it for any reason other than size minimization?

@evan
Mastodon supports only a tiny subset of the ActivityPub vocabulary. (e.g. Post payload types are limited to either "Note" or "Question" (Poll).)

* Should new clients/services avoid ActivityPub vocabulary not supported by Mastodon?
* What are we missing as a result of Mastodon's incomplete support for the ActivityPub vocabulary?
* Is it useful to propose additional ActivityPub vocabulary if it isn't supported by Mastodon?

https://www.w3.org/TR/activitystreams-vocabulary/#object-types

@Gargron
Can you say anything about the causes of the instability? Is it simply that there is a great deal of legitimate usage, or, is it the result of DDOS attacks?

@evan @atomicpoet
It appears that PubSubHubbub became WebSub at the W3C. @Julien51 can probably tell us more.

Something like WebSub, as a service to those who explicitly want their ActivityPub content to be indexed, would be a useful function for a server to provide. Such a system should dramatically reduce the cost of discovery and polling -- which is, of course, why PubSubHubbub was created in the first place as an advance beyond the old FeedMesh.

https://www.w3.org/TR/websub/

@evan @atomicpoet
Google already indexes much of the Fediverse. Below, you'll find a link to Google results for my own Mastodon handle.

It seems to me that public indexing of public data is inevitable and that public indexing of ActivityPub data would be a good thing for many applications. (Assuming that AP is more than Mastodon.) I do think, however, that more effort needs to be put into providing mechanisms for non-public, private communications in the Fediverse.

https://www.google.com/search?q=bobwyman%40mastodon.social

@Gargron 's plan to modify Mastodon to disable replies strikes me as profoundly misguided and likely to make fighting disinformation and lies more difficult.

No one who makes a public statement should have either the right or means to constrain replies -- whether positive or negative. If it is within one's right to speak, it is certainly within another's right to respond.

@evan
I think it depends on what you mean by "uninteresting."

I am quite happy to follow @GovTrack that publishes notices throughout the day about the progress of bills in Congress, etc. Others would think such posts are insanely boring.

If I had a bot that posted dozens of earthquake reports each day, that would be uninteresting for most people, but terribly interesting for a small number of folk.

What is boring for one is often considered essential by another.

"Mastodon" is too specific and "Fediverse" too technical. How about "Social Web?" As in:

* I have a Social Web account.
* I saw it on the Social Web
* Let's see what the Social Web thinks!
* Publish it on the Social Web.
* He's been on the Social Web all day.
* I'm building a new Social Web service.
* The Social Web is federated.
* Mastodon is a Social Web service and application.
* What's the Social Web's future?
* ActivityPub is a Social Web protocol.
* Will the #SocialWeb end civilization?

@evan
Back in the 1980's, I was Product Manager for DEC's ALL-IN-1 office automation product. At one point we had 45+% marketshare and were the most commonly used commercial email client. However, a key element of ALL-IN-1 was that it could be extended and customized. Thus, an ALL-IN-1 email client could use just about any email protocol -- or all of them.. As a result, there were many companies that kept ALL-IN-1 running well into the 1990's, even if only to handle email system interchange.

@evan
Some of those messaging systems taught us nothing. They came and went, leaving no trace. But, ASN.1, an X.400 component, lives on in every mobile phone on the planet, in cryptographic certificates, etc. ASN.1 inspired Googles' protobufs and now CBOR. Even JSON looks like ASN.1's textual representation.

Don't be too quick to dismiss the value of the battles we fought back in the 80's.

@evan
Someday I'd like to meet Alice and ask her what she thinks our relationship really is...

Bob

Brad @delong asked:

"We need to have an accurate theory of what caused the Fall of the Blogosphere in the first place. Why did the audience—and, yes, the creators too—succumb to the Siren song, and wind up doomscrolling through clickbait so that their glued-to-the-screen eyeballs could be sold to advertisers convincing them that the worse is actually the better product as they sell their fake diabetes cures and crypto grifts?"

@ricmac

https://braddelong.substack.com/p/old-style-blogging-should-be-new

Carl Sagan's 1990 talk on climate change is just as relevant today as it was 32 years ago.

@drvolts

https://www.youtube.com/watch?v=nno1gkceiKQ