Blast from the past

Post #443,708 by CRConrad 12/1/23 6:44:56 PM 12/1/23 6:44:56 PM Reply	Blast from the past Take a good close look at that Web page at 02:27 in https://www.youtube.com/watch?v=UBQcGXnd7rA ... -- Christian R. Conrad The Man Who Apparently Still Knows Fucking Everything Mail: Same username as at the top left of this post, at iki.fi
Post #443,714 by pwhysall 12/2/23 3:29:01 AM 12/2/23 3:29:01 AM Reply	Good eye! I’d forgotten all about that stuff.
Post #443,716 by crazy 12/2/23 7:13:46 PM 12/2/23 7:13:46 PM Reply	I knew it The second I saw that ugly shaded grid on the right hand side with that really bad text on white layout. It was tabilizer. I knew it.
Post #443,718 by static 12/3/23 1:00:49 AM 12/3/23 1:00:49 AM Reply	Interesting video.
Post #443,721 by CRConrad 12/4/23 10:39:11 AM 12/4/23 10:39:11 AM Reply	Then again, all these years afterwards... ...I find myself as sad about Bryce as I am about Norm (who also recently popped up in the periphery of my vision). Bryce was basically wrong about OOP being totally shit, yes; of course he was. But he wasn't totally wrong: OOP as we knew it back then, as we defended it at the time, wasn't necessarily all that great. At least not if the vast consensus of programmers internationally is to be believed, judging from where the debate between functional and object-oriented programming proponents is at today. And the consensus within the OOP community has since then tilted massively away from inheritance -- which AFAICR was what most of us, not just me, tried to enlighten him on the cat's-whiskerness of -- to composition. Not to mention (but I am, aren't I?) the massive conflation between OOP in general and Java's not necessarily bee's-kneesiest implementation thereof that bears the brunt of the blame for souring the aforementioned international programmers' consensus on OOP as a whole. We never got around to explaining to him that they're not the same thing either, did we? (Though I've pretty much given up on getting through to that "IPC" on that one...) And control tables are a pretty darn nifty way of setting up quite a lot of things, I thought so at the time and I still do. (Personally, I'd of course set them up as objects in my code for other objects to link to [or contain, in modern-OOP composition-speak].) He was wrong, but I can't help think he might have been salvageable somehow. Poor Norm, it seems, hasn't worked since those old days; he's on disability, if I understood correctly. He weighs in on programming discussions now and again... Mainly on subjects like DOS and BASIC. The most recent production coding I've seen him refer to having done is still in VB6. Not that there's anything wrong with that: Heaps of well-designed well-functioning profitable productive apps were built in it. If only he had been able to transition to Delphi, he could then have switched to Free Pascal / Lazarus, and for all we know he might have been productively employed all the time. (Bryce too, BTW: FP/L has, like Delphi, pretty great DB connections and APIs. Those control table apps I mentioned building above, I was totally thinking FP/L is what I'd build them in.) It's all quite sad, when you think about it too much... Though not so sad I'd want the French Mage back. -- Christian R. Conrad The Man Who Apparently Still Knows Fucking Everything Mail: Same username as at the top left of this post, at iki.fi
Post #443,724 by drook 12/4/23 3:24:05 PM 12/4/23 3:24:05 PM Reply	Basically yes Bryce took a strong position that OOP was evil and tables solve everything. He made it very hard to argue a middle-of-the-road view on the other side. Java was a great example of both an implementation of OOP, and the pathology of taking it too far, eg: with factory factories. (Is that a joke? I was never quite sure.) I've been away from the pointy end for long enough that I can't speak to inheritance vs. composition, but your larger point that different things are good for different problems makes sense. I've always been a bad evangelist for languages or methodologies, because I really care about why we're doing it this way and finding the right tool for the job, rather than just swinging the hammer harder. -- Drew
Post #443,729 by crazy 12/4/23 3:58:13 PM 12/4/23 3:58:13 PM Reply	I saw both ways too Table driven programming was the cornerstone of my existence. I created tables to drive the logic. Then I handed the tables off to the junior people to maintain. That meant I had to create an interpretive engine for the logic but it worked for me. But then I saw things that were OOP intrinsic. IBM built a printer stream language based on OOP inheritance. It matched the language perfectly. So then my brain fell into applying OOP to the problem. Then it enabled me to think that way. I could see where it was incredibly productive for this particular problem. But not most of my problems. Most of my problems were straight code or table driven. But in occasional problem really loved an OOP approach.
Post #443,752 by static 12/6/23 12:39:13 AM 12/6/23 12:39:13 AM Reply	Yah. Java is object-ish at best. I learnt more about how OO programming is supposed to work in a short Scala course than years faffing around in Java. Java is somewhere between "object-oriented" and "object-aware". And then if you want to look at a real object-oriented system, your best option is Smalltak. Wade.
Post #443,753 by pwhysall 12/6/23 2:38:59 AM 12/6/23 2:38:59 AM Reply	...oooOOOooo... No idea what you’re talking about. Now, read me in my posts.

Welcome to IWETHEY!