vkazanov

39 ms·

451.
by vkazanov 4y ago
Some time ago we migrated a few hundred thousands lines of Py2 spread over dozens of services. Easiest case possible: every service can be converted and tested separately. Took a calendar year of ca-a-a-arefully migrating these one by one,
452.
by vkazanov 4y ago
Nope. Nothing but industry-wide inertia is so massive by now that it just doesn't make sense to switch. Nosqls tried hard and went nowhere. What i find funny is that most dbs translate sql into an internal representation that is remark
453.
by vkazanov 4y ago
PS but stored procedures and udss are evil, yes
454.
by vkazanov 4y ago
It's totally possible to validate data before pushing to a table, and have normal tests around sql massaging it. Saying this as somebody looking at 1000s of transformations done by my teams. Admittedly, It took a while for data enginee
455.
by vkazanov 4y ago
As a data platform manager in a data-rich company I agree: programmers tend to prefer their favourite hammers to sql, even if a modern relational database can do the thing much better. But! As somebody with a relatively good understanding o
456.
by vkazanov 4y ago
I have to admit: no idea as well.
457.
by vkazanov 4y ago
This is not transliteration. It would be "pyotr" then. This is just using a German letter in an English text based on a purely visual similarity to a Russian letter. This is wrong on so many levels...
458.
by vkazanov 4y ago
One more thing that changed is a platform situation. Everyone is looking into building crossplatform tools, and MS tools do not really cover this front. Thus the abundance of those browser-wrappers such as Electron. The html/js stack i
459.
by vkazanov 4y ago
Did it? I remember it was something Adobe, suspected Adobe Air. Anyways, the client ui team definitely used Actionscript, which I had to read from time to time as a server-side dev.
460.
by vkazanov 4y ago
I remember how the last VisualBasic programmer left the first company I worked for. On one hand, he was ignorant in regards to everything we thought was mandatory for development: version control, code style, code structure. We, young Cpp&
461.
by vkazanov 4y ago
World of tanks used Flash-based tech for game UIs as well. PS At least that was the case 9-10 years ago when I was involved.
462.
by vkazanov 4y ago
Amazing! An incredible amount of effort! This was holding back ocaml almost for as long as I remember - and now it's just gone. Congratulations to the team! What are the next plans for the project? Spread newly available features..?
463.
by vkazanov 4y ago
If you're talking about the older approach to implementation CPython used to take then I must say if was pretty efficient for what it did (all while keeping the code clean). It took a 2-3 days deep dive for a moderately prepared progra
464.
by vkazanov 4y ago
Clickbait title mixed with confirmation bias...
465.
by vkazanov 4y ago
True. We don't. But there no "just including" in legal speak. "Including" means some sort of copyright transfer or a license. Either way, the text was created specifically for inclusion in the game and was payed for
466.
by vkazanov 4y ago
Mick's story is... Big game development in its worst.
467.
by vkazanov 4y ago
Cannot not add a quote: "victory has a thousand fathers, but defeat is an orphan." Would the author even bother with writing this post if not for game's success? Would he go to court? General public?
468.
by vkazanov 4y ago
Well, they payed for this text to be included in the game (for further distribution, obviously). That's a bad legal definition. It needs a clarification which usually comes in a form of a contract. Mojang clearly acted un-professionall
469.
by vkazanov 4y ago
He payed, and the writer accepted. That's it. It is a reasonable amount of money for the kind of work that was completed. Both sides didn't clear up the legal side of things, yes, but it doesn't change the fact that both were
470.
by vkazanov 4y ago
Knowing some superartistic and talented people, I generally agree. But Mick's case is very different. Mojang did pay, the guy did write, it's just that the guy wanted more than there was there in the informal agreement. Mick Gor
471.
by vkazanov 4y ago
...decades of research dedicated to handling problems introduced by C and amplified by C++. Sometimes it makes sense to simply reset things.
472.
by vkazanov 4y ago
The comment I replied to doesn't sound too informed. Fortified wine culture sounds more like Portugal to me, even though they definitely make it almost everywhere. EDIT: typo
473.
by vkazanov 4y ago
Hm. Russia and Balkans and the austro-hungarian region preferring sweeter wines is a centuries old culture and a wine making tradition. The same goes for Georgian wines. Vodka with wine... is that even a joke?
474.
by vkazanov 4y ago
Haskell is relatively popular of the lazy crowd. Eager Ocaml is not thriving but is definitely very alive. But Standard ML is the cleanest of the bunch, and creators kind of suppressed its development until it faded away completely.
475.
by vkazanov 4y ago
Don't do this. While Sipser has a couple of hipccups he is a MUCH better teacher and writer. Ullman et al are brilliant and all but their writing is both boring and incomprehensible. I am not even talking about the most useless recomme
476.
by vkazanov 4y ago
Have you ever seen Java/Clojure stacks? :-) My browser is cpp program. Say, 10 layers are typical. Then, there's a libc + ui linux code. Usually in c, so even less layers. Say, 10 more. Then, there are sycalls annoying the kernel
477.
by vkazanov 4y ago
strict answer: they are very deep. informal coffee break answer: abstracted layers of abstracted abstraction within parametrized abstractions just don't feel right to lispers.
478.
by vkazanov 4y ago
One (non?) problem with Scheme is that it is has a long history of being a language implementation research device. This is why it is relatively easy to implement, and this is why there are so many implementations. At least it didn't g
479.
by vkazanov 4y ago
TBH, I just speculated. I remember 60s ibm research publications on cfg, optimizations against it and was pretty sure these things are very hard to do on an ast or those simpler line by line compilers popular back then. In fact, even the or
480.
by vkazanov 4y ago
I think advanced fortran compilers already used IRs by late 60s, early 70s. Most compilers in the 80s already used some form of non-ast IRs: Generalised portable instructions, cfg representations, etc. I'd say that back then people we

More ›