galenmarchetti

10 ms·

61.
by galenmarchetti 3y ago
Yes - a ton of people. "No code" is great, tons of small businesses use them to put together internal tooling and reduce operational costs in a way that previously would only be possible with an engineering team.
62.
by galenmarchetti 3y ago
This is great to know! I wasn't aware - I appreciate the attentive response here specifically tailored around Starlink's class of capabilities.
63.
by galenmarchetti 3y ago
Feels like the elephant in the room here is...why can't the US military deploy Starlink capabilities themselves? Is this not meaningful defense technology to own themselves / worthy of legitimate defense research+spending? Feels l
64.
by galenmarchetti 3y ago
"One X user responded by saying that their small business missed out on more than $2,000 worth of sales due to the outage, during what is the biggest month of the year for the company" Ouch...that's sad. And they say cause of
65.
by galenmarchetti 3y ago
I wonder how much of the stability is the extreme reluctance to do any upgrades or touch anything related to that mainframe code. In my experience with larger institutions using mainframes, its largely "code freeze, do not mess with it
66.
by galenmarchetti 3y ago
I think, if Terraform was licensed under AGPL, then not only would all of those folks would have to also open-source their products, but also that would be the intent of the license in the first place
67.
by galenmarchetti 3y ago
Love this framing of the license...as Stallman said, who does that server really serve? https://www.gnu.org/philosophy/who-does-that-server-really-s...
68.
by galenmarchetti 3y ago
I imagine so...I wonder if its possible to draft an open-source license that commits future versions of the project to be drafted under the same license. Maybe AGPL does this (?) I know AGPL applies to all "modifications", I wonde
69.
by galenmarchetti 3y ago
didn't think about the documentation but that's especially true...it's a massive amount of copy, between all the tutorials, how-tos, educational material, and example code, all mindfully presented in a useful way
70.
by galenmarchetti 3y ago
definitely...and they get to keep that credibility and coverage. at least, as long as this move didn't totally alienate the folks that got on terraform back then
71.
by galenmarchetti 3y ago
ah wow...I guess that begs the question, "will this just blow over". As far as I can tell Elasticsearch continues to do well and is more or less in the clear (could be wrong)
72.
by galenmarchetti 3y ago
this is a great point...further down the comments, an OpenTF maintainer made a great point that all of these services are "backend hostable" services. Terraform doesn't even have a backend service component...Terraform Cloud
73.
by galenmarchetti 3y ago
would love to know a definitive/quantitative answer for #2, for sure. I do think it makes a difference when it comes to the "backstabbing" implications of the change
74.
by galenmarchetti 3y ago
wow, didn't know about this. very cool + would be interesting to see this in more places. Just checked it out. for those lazy: "If, on the Llama 2 version release date, the monthly active users of the products or services made av
75.
by galenmarchetti 3y ago
this is a great point. I've been following your work decently closely but this is the first time I thought about the "hostable" vs "non-hostable" differentiation. And thinking more deeply on it, of course it matters
76.
by galenmarchetti 3y ago
True...it's just difficult to draft a license that says something along the lines of "If you're Big Tech, we have the right to go after you, if you're small startup, you're safe". I think there's a swath o
77.
by galenmarchetti 3y ago
haha...fair point! One thing that I've been curious to see, and not sure if anyone has done it yet, is a quantification of contribution to Terraform, comparing amount of work done from external contributors to the work done by Hashicor
78.
by galenmarchetti 3y ago
100% agreed with you...would be safest for everyone to take this stance going forward. I do think, and I could be wrong, that not everyone realized that the licensing worked this way. Or had their guard up when contributing. I think that&#x
79.
HashiCorp did it backwards(galenmarchetti.substack.com)
125 points by galenmarchetti 3y ago|127 comments
80.
by galenmarchetti 3y ago
I think this whole process has been beautiful. Hashicorp was well aware that licenses are tagged to versions of projects rather than the "project" itself, and used that to its advantage as a business to start maximizing profits to
81.
by galenmarchetti 3y ago
I get what you're saying, but on a strictly legal standpoint, everything Hashicorp is doing is by the books...all previous versions of Terraform will remain under the old license and everyone can still use Terraform without fearing a f
82.
by galenmarchetti 3y ago
Creativity follows a different pattern than highly repeatable line work. Software engineering is an inherently creative pursuit, especially when paired closely with a dynamic product function. Definitely recommend Rick Rubin's book The
83.
by galenmarchetti 3y ago
I always wondered what would have happened to Bird/Lime if COVID hadn't hit. I love those things
84.
by galenmarchetti 3y ago
Crazy to think when they started, their product was a chipboard. You had to put the video monitor on yourself. What has gotten them huge, and sustained them throughout the years, was a series of products that really look nothing like their
85.
by galenmarchetti 3y ago
this is absolutely incredible. if you go to Ling Ling on Reforma, you'll get a view from the 56th story of Chapultepec Uno and you can still see how the main roads correspond to the causeways in these reproductions. you can almost see
86.
by galenmarchetti 3y ago
If you're interested in consensus algorithms, you might be interested in this book that I used in a theoretical course on distributed system called "Reasoning about Knowledge" ( https://mitpress.mit.edu/9780262
87.
by galenmarchetti 3y ago
Consensus algorithms underly distributed databases like Cassandra, or distributed caches like Redis. (in fact Redis does use a version of Raft). Simple use case, you're running a database across three servers, so that users can ping an
88.
by galenmarchetti 3y ago
unit tests generally have a faster iteration loop than end-to-end testing. so while you don't get the full coverage of a workflow, you do catch some class of bugs earlier than you would if you only rely on end-to-end testing. IMO both
89.
by galenmarchetti 3y ago
not sure it can be reduced to a single decision, but losing Ilya Sutskever to OpenAI/not cultivating his talent to its fullest extent is a pretty big "one that got away" moment
90.
by galenmarchetti 3y ago
Agreed...the user quality has fallen so much due to the advertisers preference. But it's also hard to imagine how they could have built such a long-lasting business without monetizing via ads...what would be another route to sustainabl

More ›