ehurrell

8 ms·

1.
The Reranker Tax: When a Smart Layer Can't Save a Weak Foundation(eoinhurrell.com)
4 points by ehurrell 2mo ago|0 comments
2.
by ehurrell 11y ago
Great advice. Also extends to teams, 'What do we all know well enough to execute?' should trump 'What would be fun?' every time. I've worked in situations where someone saying 'Oh I decided to write this in Clo
3.
by ehurrell 12y ago
I've worked with people who burned out and had to take time off days before deadlines doing demonstrably less than others who worked 9-6 and made sure it mattered. Delegation is certainly part of the problem, but I think there's a
4.
by ehurrell 12y ago
Instaparse is great. I've used it just enough to know how powerful it is, in a weekend project to write an org-file parser.
5.
by ehurrell 12y ago
In situations like that, much like in real life, it's prudent to have evidence (comments, emails, server logs, whatever) to protect your reputation and show the facts, but that isn't necessarily the same as having these things air
6.
by ehurrell 12y ago
The ELK stack is great, powerful, relatively easy to use and set up and very configurable. I'd definitely consider it for any dashboard prototyping, given that I've worked places where the 'MVP' dashboard could've b
7.
by ehurrell 12y ago
Papers We Love is excellent, some great papers. I forked the repo a while ago, was considering hosting one in Dublin. I think here at least the community seems to have a lot of non-CS people who I'd love to hear from too.
8.
by ehurrell 12y ago
I think the problem of staying on a version of Photoshop has existed for a long while. I agree Creative Cloud is a big misstep, but surprisingly to me some artist friends really like it.
9.
by ehurrell 12y ago
Illustrator seems widely used in brand and web design alright, but for example concept artists and comic artists I know never leave Photoshop, start to finish.
10.
by ehurrell 12y ago
Photoshop definitely wasn't intended for the wide variety of use-cases it now supports, but it is definitely the front-runner for digital painting, 95% of concept artists/comic colourists or similar artists I know or have seen use
11.
by ehurrell 12y ago
I'd agree that it's a precarious hold, but when it's lost I expect it'll be because Adobe did something, artists and designers I've talked to aren't interested in alternatives, or if they are they keep using ph
12.
by ehurrell 12y ago
While I think Krita looks really great what I've found is that artist and designers are some of the most entrenched software users around. The sentiment "you'll want to use Photoshop, because it's the industry standard&q
13.
by ehurrell 12y ago
Absolutely agree with the first part, have seen this too often, considering it's such an obvious mistake to make. People end up making an 'MVP' that is a whole-cloth beta of what they think people will want, releasing it and
14.
by ehurrell 12y ago
I agree, although with enough users this subjective marking effect can be somewhat mitigated. I'll give an example: Steam allowed people to tag games recently, which lead to a lot of 'hardcore gamers' tagging games they didn&
15.
by ehurrell 12y ago
Is this the entire story? It's true that there are many times more iOS users than OSX users, but what percentage of each are their target market (who I would call 'power users')? Panic make great apps (I use them on both iOS
16.
by ehurrell 12y ago
I agree with you here, but I think 'stable libraries' is perhaps a good target for a few reasons, right now the culture isn't just bad code, it's "There is no advantage or benefit to showing your code". I would
17.
by ehurrell 12y ago
I think it might be significant, or at least hint at a significance worth exploring. Taxi drivers' brains apparently adapt and grow a much better awareness of maps and routes than the regular person, and this might be the same thing ha
18.
by ehurrell 12y ago
One of the best messages I got from the How To Start A Startup course was an off-hand line during the funding class: the best thing you can do for your funding chances is have a great product, and Mixpanel seem to nail this, great stuff. As
19.
by ehurrell 12y ago
Especially with games, gameasutra runs postmortems with devs frequently to talk about both successes and failures in their projects.
20.
by ehurrell 12y ago
It seems to take all kinds. If the other party wants to blacklist you they'll usually find a reason to allow themselves to, though usually people are reasonable enough that things don't get that bad.
21.
by ehurrell 12y ago
Absolutely, it's hugely important. The core idea that subjective beings will disagree holds obvious weight, but it takes serious commitment to improving the process to offer up this kind of experiment to prove it.
22.
by ehurrell 12y ago
I've had similar experience, and it can be tough, especially when the lack of interest from a reviewer is obvious. I still joke about the review I got that included the line "English need improving".
23.
by ehurrell 12y ago
Having done a CS PhD it very much depends on the field. Yeah you'd be laughed at for running an algorithm to find out run time, but areas like information retrieval tends to involve humans and their reactions enough that experimental d
24.
by ehurrell 12y ago
One of the most difficult things to learn as a CS PhD is that you aren't an engineer, you are a scientist, so you spend your time learning to approach problems in rigorously scientific ways. In some ways it's a career switch. Doin
25.
by ehurrell 12y ago
I continue to feel extraordinarily lucky about my experience with my PhD. It was grueling, it was tough, but I had some fantastic support. I know many who weren't so lucky, and the effects are profound.
26.
by ehurrell 12y ago
From the commentary: "we are always more accepting of an expected outcome", which I think says a lot. Random Forests are great, and work well, but sometimes it feels like there's a world of difference between Kaggle-contest-s
27.
by ehurrell 12y ago
Thanks for the resources, my current hobby project is attempting to recreate deep learning in Clojure, have been burning through material, cheers!
28.
by ehurrell 12y ago
I've never used it, but the way it came to my attention might be telling; hearing of a startup thinking people were posting about them on Secret. A 'pay what I want to reveal who I am' model could be interesting, rather than
29.
by ehurrell 12y ago
Nice! I'm continually impressed by the Shiny stuff I've seen, very quick to make something impressive.
30.
by ehurrell 12y ago
I was wondering why I hadn't seen ye about :P Irish people going abroad to start companies seems to work well, good luck :)

More ›