2022-05-05 00:14:07 <KipIngram   > lispmacs[work]: Looking at those "subsistence in terminal mode" section of your Gemini site.
2022-05-05 00:14:33 <KipIngram   > I remember those ctrl/alt <number> terminals, but I haven't seemed to be able to get them to work the  last few times I've tried.
2022-05-05 00:14:51 <KipIngram   > If I hit ctrl-alt-<num>, I get... nothing.  Completely ignored.
2022-05-05 00:19:24 <KipIngram   > Not surprised re: PCB layout having no ready solution.  That's always struck me as a "must have graphics" task.
2022-05-05 04:01:20 <dave0       > this is a nice quote from chuck moore... https://www.red-gate.com/simple-talk/opinion/geek-of-the-week/chuck-moore-on-the-lost-art-of-keeping-it-simple/
2022-05-05 04:01:29 <dave0       > RM: Are programmers and computer scientists aware enough of the history of technology? It is a pretty short history after all?
2022-05-05 04:01:29 <dave0       > CM: I lived through the history of computers. And I missed a lot of it. History is a brooding study that adds flavour to a subject.
2022-05-05 08:05:28 <KipIngram   > Any of you guys watch any of the seres "Halt and Catch Fire"?
2022-05-05 08:05:38 <KipIngram   > s/seres/series/
2022-05-05 08:06:38 <KipIngram   > I thought the first couple of seasons were pretty good.  Captured the "feel" of some of my small company days really well.
2022-05-05 08:07:00 <KipIngram   > Then they sort of moved on into other aspects of the history of the computer industry I never had anything to do with, and I lost interest.
2022-05-05 08:09:42 <KipIngram   > One bit that just kind of rolled my eyes, though, was when the two guys were dumping the BIOS ROM of a PC, they wrote the numbers down in a notebook.
2022-05-05 08:10:20 <KipIngram   > Come on - you know they'd have typed them into a second computer instead.  They had to get from that notebook into a digital form anyway - why insert that extra arduous step?
2022-05-05 08:10:53 <KipIngram   > That was just one more layer for errors to occur in.
2022-05-05 08:28:56 <KipIngram   > You know, I don't think I really realized until recently here, as I did that UTF-8 work on EXPECT, how much of my system is "there" to support reading user input from the keyboard.  That would be far and away the largest "one purpose" chunk of code in the system.
2022-05-05 08:29:09 <KipIngram   > I guess that makes sense - that's where you have to deal with all the "user randomness."
2022-05-05 08:32:49 <KipIngram   > It also makes it pretty clear why the Linux folks bundled read up the way they did.  Handling all that business of the user tinkering around with keys, and just sending the finished line back to the user - that offloads quite a lot of work from having to be done in more or less every single program.
2022-05-05 13:23:42 <KipIngram   > Well, this is kind of a punk day.  I can ssh into all of my lab boxes at work, but Outlook, Trello, Slack, Box, and WebEx are all broken.
2022-05-05 13:23:49 <KipIngram   > Something unhappy is going on.
2022-05-05 13:23:56 <KipIngram   > I'm effectively incommunicado.
2022-05-05 13:24:05 <KipIngram   > On the work front.
2022-05-05 13:24:14 <Zarutian_HTC> really?
2022-05-05 13:24:25 <KipIngram   > Yeah, really.
2022-05-05 13:24:38 <Zarutian_HTC> and you can finally get some work done?
2022-05-05 13:24:55 <KipIngram   > Hah hah heh...
2022-05-05 13:24:58 <KipIngram   > Exactly.
2022-05-05 13:25:21 <KipIngram   > Part of me is annoyed; part of me is kind of liking it.
2022-05-05 13:25:27 <Zarutian_HTC> lesse, most of these are ms azure cloud based?
2022-05-05 13:25:55 <KipIngram   > Potentially - I'd have no idea.  But it does seem reasonable that there's some shared resource that's busted today.
2022-05-05 14:12:40 <KipIngram   > Darn.  Things seem back online now.  ;-)
2022-05-05 14:42:11 <Zarutian_HTC> why is it that many work places insist on interruption prone stuff?
2022-05-05 14:55:26 <KipIngram   > A lot of things they do mystify me.
2022-05-05 14:57:35 <KipIngram   > Like the sheer intensity that gets put behind "agile development."  I don't really have anything against agile in principle - there are some sensible practices woven into it.  A lot of them I "kind of did" before I ever heard about "agile."  But the damn near religious fervor that gets put behind it, and the latent attitude that "if we follow this process we'll win" I just don't get.  Sure, organization is
2022-05-05 14:57:37 <KipIngram   > good.  But a particular *style* of organization and discipline isn't going to suddenly inject creative genius into a group of people.  The talent has to be there, and has to be allowed to operate.
2022-05-05 17:30:42 <veltas      > KipIngram: I think everyone is religious, some people just don't realise it
2022-05-05 17:31:06 <veltas      > Iterative development seems like a good methodology to me
2022-05-05 17:32:06 <veltas      > There is a huge practical difference between waterfall and an iterative approach
2022-05-05 17:33:00 <veltas      > I also think a lot of software engineering process can be replaced by pure code and code tools
2022-05-05 17:33:59 <veltas      > And it's easier to explain to someone fresh out of uni the differences when you just talk about it as code
2022-05-05 17:34:55 <veltas      > You say "okay, you've got a big app planned, you want to write it top-down, but you're finding it hard to figure out what the parts are and what the data looks like..."
2022-05-05 17:35:02 <veltas      > "that's design"
2022-05-05 17:35:26 <veltas      > "You go to start some functions etc that you know you need and the work quickly spirals out of control... that's waterfall"
2022-05-05 17:35:39 <veltas      > "let's start with something we know we can do... that's iterative"
2022-05-05 17:36:32 <veltas      > "but my boss wants to know when we'll be done!" "tell him it's 'agile'"
2022-05-05 17:41:02 <veltas      > "Where's your design document?" "main()." "Where's your tech spec?" "main()." "Where's your control drawing?" "git"
2022-05-05 17:41:03 <remexre     > yeah, I work in defense contracting, and we've been required to certify that we're doing scrum before...
2022-05-05 17:41:21 <veltas      > The only doc I want to actually write is a manual usually
2022-05-05 17:41:31 <veltas      > I also work in defense
2022-05-05 17:41:52 <veltas      > Well defense is the main customer of the business I work for, rather
2022-05-05 17:42:21 <remexre     > yeah, same
2022-05-05 17:42:58 <remexre     > there was actually a really funny doc from... darpa? a few years ago
2022-05-05 17:43:01 <remexre     > "spotting agile bs"
2022-05-05 17:43:21 <veltas      > We do ISO 9100, the defense version of that, we put 'agile' and 'scrum' in our QMS (I think a lot of contracts probably ask for that in writing... I think there's public DoD white papers on that)
2022-05-05 17:43:22 <remexre     > a guide for PMs trying to figure out if they were _really_ doing agile or not
2022-05-05 17:43:28 <veltas      > Yes
2022-05-05 17:43:35 <veltas      > That's one white paper I've seen
2022-05-05 17:45:49 <veltas      > One problem is that agile is not appropriate for many problems
2022-05-05 17:46:40 <veltas      > i.e. if you're designing a computer, then you CANNOT use agile. There is no "MVP", when the computer has been made, it's been made, with or without the features you wanted
2022-05-05 17:48:03 <veltas      > Although the designs can certainly be iterative, you can't do iterations in 2 weeks
2022-05-05 17:48:21 <veltas      > And you can't "break early"
2022-05-05 17:49:39 <veltas      > I noticed a team at work was actually doing something that resembled agile recently and me and a few other engineers remarked how "wow you are actually doing agile"
2022-05-05 17:50:50 <KipIngram   > veltas: Yes, I was trying to be careful not to criticize the actual work that agile encourages.  Sometimes it can be quite smart.  I just think the "smart" comes from having capable people on the team, that have some design experience, and not from the particular process being run around it.
2022-05-05 17:51:19 <veltas      > My point is that the process does actually matter, but also yes it's not smoke and mirrors
2022-05-05 17:51:20 <KipIngram   > I definitely prefer iterative over waterfall as well.
2022-05-05 17:51:52 <veltas      > scrum is like iterative on steroids, and I have concerns it's crushing and might cause burnout, but not a ton of experience with it
2022-05-05 17:51:55 <KipIngram   > I think some sort of organizing process that exists, that allows the right kind of work.
2022-05-05 17:52:06 <KipIngram   > "Chaotic design" rarely pays off.
2022-05-05 17:52:28 <KipIngram   > Usually when it does it's because someone on the team HAS organized things, just without making a show of it.
2022-05-05 17:52:55 <veltas      > Often I think arbitrary structure is arbitrary and adds cost
2022-05-05 17:53:05 <KipIngram   > ^ I agree.
2022-05-05 17:53:35 <veltas      > People do need to understand engineering, planning, design etc, and I doubt many orgs have created a 'process' that suffices where engineers don't know how
2022-05-05 17:53:57 <KipIngram   > Yes!  That's what I was trying to say.
2022-05-05 17:54:24 <veltas      > But agile/scrum has a lot of good ideas, some of it is quite cargo cult, but it's not universally applicable
2022-05-05 17:54:59 <KipIngram   > Yes.
2022-05-05 17:55:18 <veltas      > When I spend a month trying to fix a nasty firmware bug that wasn't because I "wasn't doing scrum right". I am just fixing one thing, there's no way of dividing that up into a 2 week objective. I could have taken 2 days to fix it but it took me a month
2022-05-05 17:55:26 <veltas      > And sometimes I think I'll take a month and it takes me an hour
2022-05-05 17:56:16 <veltas      > And when I've got a shitty scrum methodology in the background asking me every day what progress I made and what's 'blocking' me it's actually a big waste of my time and a bit demoralising
2022-05-05 17:56:17 <KipIngram   > Just the focus on getting people "unstuck" is a good one - you'd think people would be very motivated to get unstuck and would get that done, but I think sometimes it's easy to let it go on too long.
2022-05-05 17:57:06 <KipIngram   > Yes - it all depends on how it's handled.
2022-05-05 17:57:47 <veltas      > If your standup turns into "show and tell" you've done it wrong. And I'm guessing 99% of companies do it that way
2022-05-05 17:57:53 <KipIngram   > Sometimes not having progress to share doesn't mean you're stuck - sometimes it just means you're chasing a goal that takes more than a workday to reach.
2022-05-05 17:58:19 <KipIngram   > And I don't believe you can ALWAYS break things down into pieces that small.  Sometimes tasks are hard.
2022-05-05 17:58:38 <veltas      > Yeah
2022-05-05 17:59:06 <KipIngram   > Probably my biggest opinion here is that you shouldn't start trying to put that process in place until you've got a clear product architecture that's been well-validated.
2022-05-05 17:59:28 <KipIngram   > Up front planning doesn't always produce a lot of showy results, but it's the most important part of the whole process.
2022-05-05 18:00:50 <veltas      > Yeah it's true, there's a lot of up-front risk
2022-05-05 18:00:59 <veltas      > And it usually gets totally shoved under the rug
2022-05-05 18:01:24 <KipIngram   > You just don't want to get 70% of the way to the finish line and realize there was something you overlooked.
2022-05-05 18:01:45 <veltas      > "It's fine, it will all work out. *chuckles*" No I don't want your sarcastic laugh, I want to know this project won't waste months of time
2022-05-05 18:02:29 <veltas      > I've got X years on this planet and I don't want to spend one of them fixing your fecklessness
2022-05-05 18:03:52 <veltas      > I do properly design stuff and really plan and think out stuff, I don't want to write crap. But then you get people who don't want to listen or don't care
2022-05-05 18:04:12 <veltas      > And then months later they walk into the exact problem you described a solution to already and it's too late
2022-05-05 18:05:07 <veltas      > This is one of the reasons iterative is so important: if you can figure out how to start simple and go step by step then you do really de-risk the project
2022-05-05 18:06:20 <veltas      > Figure out logical iterative steps and then even the most feckless engineer will write something acceptable
2022-05-05 18:06:53 <veltas      > Sorry I've been stuck on 11 all day
2022-05-05 18:08:39 <KipIngram   > No worries - all that resonates with me.
2022-05-05 18:09:58 <veltas      > Iterative is just tick-tock, it keeps it all moving and the money flowing. If the music ever stops we'll realise there are no chairs left
2022-05-05 18:10:55 <veltas      > Speaking of which this is fantastic, please look if you didn't see it already https://ciechanow.ski/mechanical-watch/
2022-05-05 18:34:17 <KipIngram   > Oh cool - I just spent a really pleasant hour or so a couple of weeks back reading the Wikipedia history of the pendulum clock.
2022-05-05 18:34:27 <KipIngram   > Time-keeping devices are *fascinating*.
2022-05-05 18:36:20 <KipIngram   > Ok, I'm saving this for the weekend.  That's gonna be fun - thanks for sharing!