To take it in a different direction, Arrigo Sacchi (one of the more legendary managers of AC Milan), who had no professional sporting background, when questioned about his qualifications said "I never realised that in order to become a jockey you have to have been a horse first."
The analogy with football players/managers was in my head too — there are many average professional footballers that go on to become fantastic managers and there are lots of exceptionally talented footballers that make terrible managers.
The contrast to this is to put in tickets specifically for refactoring and reorganization, but I've rarely seen that work since they often don't have any sweetener included to encourage e.g. product organization to sign off on the work.
Yeah, it's a shame, I often have to do this type of crap in my off time. However, having a well architected app significantly reduces these types of massive undertakings. Doing it right the first time has its advantages. Then again, weighing it against delivering early and other important aspects has its merit too.
It's just a double edged sword unfortunately that you often have to walk in this industry.
The counterpoint to that is that for the same SNR where you're at barely (or not at all) audible or garbled on analog, you have absolutely clear digital audio.
Yeah, and with most comms being stereotyped, most messages carry little data, meaning you can easily add redundancy and still take up the same or less airtime. You will still need plaintext messages for complex situations, but even that can be transmitted more clearly and efficiently with digital encoding.
At least for my workflow, which is fairly similar, the result of the first few stages is a refined single prompt, which I then run and the result is a PR. so I, at least, would include that final single prompt over the prior intermediate results.
I also think that might not be everything you need to really vet/recreate those PRs since there's a lot of strategic and operational stuff that goes into the pre-composition runs (sometimes I actually throw everything away and restart), so something a little more sophisticated would be interesting.
These days setting up a proper system on your build orchestrator of choice is pretty easy, cheaper, and you basically only lose a tiny bit of UI niceness if they report to github anyway
This is what I've been trying to do for over a decade - you can go even further with tee-testing by making a copy of the live traffic you run in a sandbox environment and match responses against. Obviously you can't do some things in the sandbox, email and payments high on that list, but e.g. payment gateway responses can be "borrowed" from production.
Once your new version has no disagreements, it's a really clean swap. I am working towards developing this way and I think, if taken to extremes and carefully, it can remove the review requirement most of the time.
I'm confused, we did this on US Tiger line file data in 2006, rendering it on a early equivalent of Google Maps. You have to decimate the polygons to get them to render but doing bounding boxes around postcodes is fairly reasonable.
I think the one I enjoyed the most was a zipcode with 38 separate geographical areas, and the other example was a street where one half was one zip code, and the other was a different one, but half way down the street they swapped sides. Representing that with non-overlapping polygons was impossible. So we just let them overlap and shrugged.
I was talking about runtime rendering. It's a bit of a hard topic to talk about, as there are a multitude of use cases and approaches.
But in general, having more points really doesn't help with processing geo data. As you said yourself, you had to decimate the polygons. It's always compute vs. precision.
Well, we did it real time using VML in IE6 of all things, and we used adaptive decimation based on the zoom level, so from a practical perspective there's no point in rendering subpixel detail levels
I've been doing software engineering for 20 years and approximately half of my total work output has been in the last 2 years. Once you get it set up and maintain it regularly it's easily three to ten times faster
I'm just going by github, which has captured the majority of my work since 2007, and says my prior peak was in the ~800 contributions per year, and this year so far I've done 5k+. Not always comparable but close enough I feel comfortable saying "at least 3x"
https://en.wikipedia.org/wiki/Arrigo_Sacchi
reply