g-clef

having worked in a government agency that ditched IBM, let me offer a view of what that looks like from the customer side:

IBM bought a company whose product we'd been using for a while, and had a perpetual license for. A few years after the purchase, IBM tried to slip a clause into a support renewal that said we were "voluntarily" agreeing to revoke the perpetual license and move to a yearly per-seat license. Note: this was in a contract with the government, for support, not for the product itself. They then tried to come after us for seat licenses costs. Our lawyers ripped them apart, as you can't add clauses about licensing for software to a services contract, and we immediately tore out the product and never paid IBM another dime.

I tell this story not to be all "cool story, bro", but to point out that IBM does focus on renewal growth, but they're not geniuses...they're just greedy assholes who sometimes push for growth in really stupid ways.

Until passkeys can pass the test of "my non-technical friends and family don't call me for help about them", passkeys aren't ready. Vendors keep making assumptions about how users behave which are not safe assumptions, and that keeps blowing up the interactions of non-technical users. (I'm sure there's an "assumptions developers make about user accounts" blog out there somewhere.)

For example, my family has had to call me for help on the interaction between passkeys on Apple & Amazon multiple times. They have a shared Amazon account, which neither Amazon nor Apple seem to like. The first problem came when they didn't even know they'd been moved to passkeys - there was a popup that one of them didn't understand, they clicked OK to get it to go away, and suddenly the other partner can't log in, and neither of them can figure out how to log into Prime Video on their AppleTV. Another time one of them got "nudged" to add a fingerprint to the account, again freezing out the other person.

Until that nonsense stops happening, Passkeys aren't ready.

It's funny, I started with rPis for the same reason, but I'm about to replace them. I bought 20 rPi 4Bs for my homelab, and I just couldn't get them to do what I needed. I was looking to run a home k8s cluster and the Pis were just not suited to it at all (don't use sd storage for k8s 'cause it'll burn out the card w/writes, booting off usb was unstable even with powered usb hubs, netboot turned into an enormous pain in the neck).

Yeaaaahhhh...NLM is/was a special beast inside NIH.

At the time I was there they had a budget to die for, and Pubmed was in the top 300 websites in the world (both are probably still true). NLM pushed so much traffic because of Pubmed that they had their own internet connection to avoid DoS'ing the commodity desktop traffic of the rest of the NIH.

I'm going to regret this, I bet, but here we go: Many years ago I was the manager of the team inside NIH/CIT that ran both the border firewalls and the DNS servers in question (to be specific, at the time it was NIH/CIT/DNST/NEB/NSS). Obviously, since I'm not there anymore I don't have any special inside information, but given that the DNS servers were responding to TCP and not UDP during the outage, my bet would be a simple firewall screwup, rather than malice.

Stuff happens - maintenances have unintended consequences, people typo stuff, etc. Don't freak out about every event - save your powder for the real outrages (like 18F).

> It was reasonable to expect that most of them didn't have the technical capacity to accomplish that in the available timeframe.

So what? Just because the owner can't respond in a given timeframe does not give a you (or anyone) the right to appropriate other people's property. By your argument at the height of the Covid lockdowns I would be justified in taking your car & loaning it out to people because you weren't using it and didn't "have the technical capacity to accomplish it in the available timeframe."

The fact that it was a license that IA was assigning to themselves rather than a physical object makes no difference whatsoever.

That can have downstream effects, though. When I talked to them about doing it to my more recent Subaru, they told me I'd lose the front speakers and the in-car microphone as well, since all of them went through the same fuse.

The title is a bit misleading. The point of the article isn't that their ethics board resigned - that happened a while ago (right after the Uvalde shooting). The point of this article is that even after the board resignation over taser-carrying drones and a shareholder proposal to stop drone development, Axon bought a drone company anyway.

Interesting. This could be a bracketing error, because I read

> it includes analysis of many older accounts that haven’t sent tweets in the last 90 days and thus, likely don’t fit

> Twitter’s definition of mDAUs (monetizable Daily Active Users)

As implying that they think accounts that haven't tweeted in the past 90 days don't fit Twitter's mDAU definition. Given the placement of the qualifying phrase, I think that's a reasonable parsing of the sentence, but I see your point that they could be trying to imply their set doesn't fit the definition. If so, that sentence is very badly constructed.

As a third party? Probably not. Which is why it's going to be very hard to disprove Twitter's assertion unless Twitter chooses to share their data.

That's part of why I find articles like this frustrating: I don't think they have the data to actually answer they question they're attempting to answer. Knowing that, what's the purpose of the article?

Except the question isn't about the pure number of spam/bot accounts, it's about the ratio of spam/bots to "authentic" users. If you leave out the lurkers, that ratio gets skewed to mistakenly inflate the bot count.

Oh, do I have notes on their methodology.

1) They talk about "active" accounts (meaning have tweeted in the last 9 weeks), and do a bunch of filtering against that. That seems like a huge bias - lurkers exist, and in my experience are usually the majority of users...this step removes them or ignores them entirely. Frankly, until recently, my twitter account would have been one of the ones they would have discarded as inactive. This one thing alone makes me question all of the rest of their results.

2) By the same token, the rate or frequency with which a user sends tweets has no relation to whether a user is monetizable. If they're seeing ads, they're monetizable...lurkers are just as monetizable as high-volume posters.