Y
HN Search
Hacker News Search
new
|
comments
|
top
|
jobs
philbo
searching PlanetScale…
1.
▲
2.
▲
3.
▲
4.
▲
5.
▲
6.
▲
30 ms
·
331.
▲
by
philbo
13y ago
It's still pretty early days with it at the moment so things are liable to change, but I have elastic set up right now. I'm completely new to search though, so I don't really know the ins and outs of different solutions. Plen
332.
▲
by
philbo
13y ago
Currently I'm working on a few different developer tools, as part of my continuing mission to make my own life easier. I figure they're probably cool because if I'm scratching my own itch then I should be scratching itches fo
333.
▲
The Karma Hypothesis
(dilbert.com)
4 points
by
philbo
13y ago
|
0 comments
334.
▲
by
philbo
13y ago
+100 Why some developers are so against progressive enhancement mystifies me. It is an elegant solution that actually works in all cases rather than an ugly hack that should probably work in the majority of cases. How can there even be a di
335.
▲
by
philbo
13y ago
Even if the 10x engineer were to exist, assuming a normal distribution, they would be extremely rare and highly sought after. To optimise your business around hiring such a precious unicorn would seem like a very foolhardy strategy. For one
336.
▲
Show HN: An email reminder service for your GitHub starred repos
(githubreminder.org)
5 points
by
philbo
13y ago
|
3 comments
337.
▲
by
philbo
13y ago
You think the only people interested in classical music and literature are "some scholars maybe"?
338.
▲
by
philbo
13y ago
I'm going to try modulus.io later on today. Obviously I don't have an opinion yet but my reason for giving it a try is that I've been told you get a free database instance in the same data center as your node instance, which
339.
▲
by
philbo
13y ago
Absolutely! But I still find myself having to make the case for it, when others are looking for ways to cut corners.
340.
▲
by
philbo
13y ago
Strongly disagree. Lots of the web may not degrade gracefully, but then also lots of it does. Further, the notion of it as a lost battle is unfortunate because we're not discussing a binary choice between two opposing sides here. As an
341.
▲
by
philbo
13y ago
The following rant is somewhat tangential but, as a front-end developer that takes pride in progressively enhancing websites I work on, I think this is a shame for a different reason. So many times when speaking to employers/product ow
342.
▲
by
philbo
13y ago
Oh, I had no idea it wasn't portable, sorry. I'm using iTerm2 on OSX, but maybe worth trying them without the colouring: git log --graph --all --abbrev-commit --pretty=format:'%d%x20%h%x20%an%x20%s' git log --a
343.
▲
by
philbo
13y ago
Here are two variations that I find useful day-to-day. 1. Graph showing commits on all branches, decorated with branch and tag names, short commit hashes, committer name and commit comment: git log --graph --all --abbrev-commit --pret
344.
▲
by
philbo
13y ago
Straw man. At no point does the parent advocate for smalltalk on HN or stamping out contrary opinion en bloc. The point is purely about behaviour, respect and being considerate of others.
345.
▲
Pa11y, your automated accessibility testing pal
(github.com)
2 points
by
philbo
13y ago
|
0 comments
346.
▲
by
philbo
14y ago
Tbh, I was speaking from a position of ignorance. The last time I used sourceforge properly must have been circa 2006/7 and it was an abomination then - clearly that is not the case now. It definitely felt cleaner from a UI perspective toda
347.
▲
by
philbo
14y ago
Judging from the `.dsw` file at the base of the code directory, it looks like you need to build it in MS Visual Studio.
348.
▲
by
philbo
14y ago
Is your wrapper open-source and available in NPM? If not, it should be! :)
349.
▲
by
philbo
14y ago
Preferring sourceforge over github seems like an unusual choice, I didn't realise anyone still used it. It's a git repo too, I wonder why they did that?
350.
▲
by
philbo
14y ago
Beyond any educational benefit, what do you see as the advantages of programming in this way? I and most of the programmers I socialise with tend to see comments as a bad smell; they're fine every now and then but if there are more than a f
351.
▲
Browserify 2.0
(gist.github.com)
2 points
by
philbo
14y ago
|
0 comments
352.
▲
by
philbo
14y ago
BIZZBY ~ LONDON, UK ~ FULL TIME Bizzby is a well-funded startup preparing to revolutionize the economy. We are a team of talented, highly-motivated developers and designers based in Shoreditch, working hard to deliver jaw-dropping software
353.
▲
Node Cluster and Express
(rowanmanning.co.uk)
2 points
by
philbo
14y ago
|
1 comments
354.
▲
by
philbo
14y ago
I'm not a fan of the format for gruntfiles (which is not to say I think it's objectively bad, just that it doesn't appeal to me personally). I much prefer using Jake as the build tool for my JS projects: https://github.com/mde/jake
355.
▲
by
philbo
14y ago
This bit surprised me somewhat: 'Not only had we gotten very used to calling it that, the string “Alexander” was all over the code and buried in its resources. So the engineering team wasn’t just curious about the real name, they were worri
356.
▲
by
philbo
14y ago
I didn't really understand what this was until I scrolled down to the screenshot and realised its a buildwall. May be worth using that term in the summary on GitHub for clarity? As for the project itself, nice work! I'm a huge fan of buildw
357.
▲
by
philbo
14y ago
I suspect there are some subtleties to that first answer which you've overlooked. It is written by Kent Beck. He pretty much invented the concept of test-driven development. He's probably the most well-known advocate of developers writing t
358.
▲
Show HN: Complexity analysis for JavaScript projects
(jscomplexity.org)
4 points
by
philbo
14y ago
|
0 comments
359.
▲
Show HN: complexityReport.js
(github.com)
2 points
by
philbo
14y ago
|
0 comments
360.
▲
by
philbo
14y ago
There's been a lot of talk [1] about promises on es-discuss in the last week or so. It looks like an effort is underway with a view to standardising them as part of the language in the ES7 spec, possibly with a harmonised DOMPromises spec t
More ›