I was able to finish reading PragProg’s book on LiveView just yesterday (At least the parts that are currently published)
I’ve got kind of mixed feelings though… the content is incredibly interesting, of course, but this is the first PragProg book that I just had trouble getting through… Maybe it’s just that it’s not polished yet due to its EAP status (I hope that’s the case!), but I don’t know, somehow the issues I seemed to have with the text are a bit more fundamental than what could be easily fixed? I constantly had the feeling like the content is being literally repeated over and over again, and felt the constant need to skip through the lengthy paragraphs of text that had no “meat” in it just to get to the actual code…
Haha, yep, my desktop is… not modern, but still blazing faster than basically every other computer I see by anyone else (boots in 2-3 seconds, launches programs instantly, etc… etc…). It’s all a Phenom ][ x6 (6 full native cores! Pre-hyperthreading, lol), 16 gigs ram (max the mobo supports), an AMD CYPRESS generation GPU, etc… Lol.
And I can afford new stuff fine, that’s not the issue, I just hate getting replacement parts until the part I’m replacing has properly and fully died, which nothing in my machine does anymore (cpu is almost 11 years old, the GPU is like 9, the hard drives range from 7 years to 25 years old, the PSU is like 20 years old, etc… etc…), I’ve been upgrading this same system part by part as things die for over 30 years, lol.
For running modern games and such I borrow my wife’s system, vega and all. ^.^
And screw nvidia, their drivers are horrid abominations that every programmer should be reviled from, lol.
EDIT: And yes, I run tests often on my system, everything is in proper working condition except for a 2meg chunk of RAM that I’ve blacklisted in the linux kernel so the system still runs, lol.
I completely get the sentiment of “don’t over-consume and try to reuse your tech as much as possible” but it seems like you might be missing another factor – power consumption. Old tech is very power-hungry and inefficient.
Reading your posts (and knowing you from before) I am getting the impression you can be extremely well-served by one of the Ryzen R1000 / V1000 / V2000 embedded systems (so basically a 1L - 2L volume of computer) and never go above 50W usage.
So that’s an extra factor to consider. My main desire to move to full Ryzen and/or Apple’s ARM is, apart from monstrous speed, extreme power efficiency.
Actually mine throttles down very well, it runs at 800mhz most of the time, no screen 99% of its life (I heavily access it remotely), etc… I do use it as a build engine among other things so it can’t ever really be turned ‘off’, but overall it’s more power efficient then any of the newer ones I use.
For bigger/modern things I just borrow my wife’s computer, she’s a heavy gamer so it’s a bit more powerful, and yes she has a ryzen 7. ^.^;
Plus, power isn’t really a big issue in my area, it’s really cheap here because I live adjacent to giant wind energy farms (and it’s always windy here) so no carbon footprint either.
Last week, I started my first real-world project in Elixir! It’s a Slack bot that when connected to your Slack workspace, listens to flag emoji reactions (such as or ) to messages (in channels that the bot is invited to) and once such reaction is added, it translates the post that got the reaction to a language indicated by the reaction and posts the translation as a reply to the original message.
Since the company I work for has bases all over the world but not everyone can speak English, this helps to solve the issues you’d normally encounter with cross-region branch communication!
Also, what was a super exciting factor for me was the fact that just after three evenings of casual coding, I had a proof-of-concept version of the bot working! Two more days of tweaking and tuning and the bot got approved and deployed to our company’s Slack workspace. From zero to production in five days!
So continuing the theme of translation bots, I decided I’ll give a crack to another translation bot - this time for the LINE communication platform. And just like with Slack, I was able to go from zero code to working bot in a matter of few days – is Elixir awesome or what?
… now just to figure out why the LINE webhook payload sometimes omits userId indicating the sender of the message, that’s quite an important piece of information I need
(At least the parts that are currently published)
or 
) to messages (in channels that the bot is invited to) and once such reaction is added, it translates the post that got the reaction to a language indicated by the reaction and posts the translation as a reply to the original message.
Anyone else attending?
