tterrace

Dos32.bas with Visual Basic was my first intro to programming. There was no useful search so you had to navigate through webrings to get programming tutorials. There were also networks of private vb channels where the hackers hung out, those were always fun to drop in on.

Maybe I’m looking back with rose-colored glasses but I remember Visual Basic being intuitive and approachable for beginners in a way that I haven’t seen since.

The fader text is a nice touch too, that immediately makes me nostalgic.

Recently there's been cases where searching for a company name will actually return an ad for a competitor. Companies are being forced into a bidding war to win back the top position for exact searches on their own name.

See: https://twitter.com/jasonfried/status/1168986962704982016 and the SS in the reply: https://twitter.com/webiyo/status/1205534848192045056?s=20

I'm not on buildbot nine (I think the new waterfall UI is a big regression), but what that is describing looks like a statically defined list of workers that scale up and down dynamically. What I'm looking for is the ability to add and remove workers at will, without having to add them to the configuration list and restart the master.

In terms of underpowered build steps, I have several fairly complicated, 1k-2k line build factories, with multiple codebases and hundreds of steps (some custom, some from the stdlib). There's many dependencies between the steps, and many different properties that can be toggled in the UI. All these steps need to be defined up-front in the master, but their actual execution often depends on runtime information, so it becomes a mess of doStepIfs. I think it would be an improvement to give a program on the worker the power to tell the service what it wants to do, rather than the other way around.

I'm still on buildbot, but it's definitely showing its age and I'm hoping to move off of it within a year. I've been keeping an eye on Chromium's buildbot replacement, LUCI (https://ci.chromium.org/). It's still light on documentation and the source is very internal google-y (they seem to have written their own version of virtualenv in go). However, based on the design docs it does look like they ran into a lot of the same problems I have with buildbot, specifically the lack of support for dynamic workers, and how underpowered the buildbot build steps can be.

https://github.com/luci/recipes-py/blob/master/doc/user_guid...

A recent example of this is buildbot. The current version (8.x) has a very no-frills, utilitarian python/jinja UI which has served me, personally, and many other open source projects well (chromium, mozilla, LLVM, etc). Version 9 is a rewrite two years in the making using angular with the modern JS toolchain for its frontend. I tried out the latest (beta) build, and from what I can see I think it's exactly what you describe: more resources with less functionality. Time will tell whether the final version lives up to its promises but I'm very skeptical at this point.

Amazon also has a cloud deployment tool called Apollo (internally only I believe - the public version is called "CodeDeploy"): http://www.allthingsdistributed.com/2014/11/apollo-amazon-de...

The lunar laser ranging experiment was used to observe the effects of relativity: http://en.wikipedia.org/wiki/Lunar_Laser_Ranging_experiment

What was the performance like for those queries?

Could you link the "why is my go program 138GB!" article?

"Put the secret into your shared/.rbenv-vars file"... and then say goodbye to them! http://www.cve.mitre.org/cgi-bin/cvename.cgi?name=CVE-2014-0... . I don't think keeping your secrets in a file accessible to the web server is a good idea because of LFI vulns like this.

Some other ideas that I've heard that may be better: store the secrets on a separate "offline" server that only the web server can talk to. Or have the file readable only by root, run a bootstrap script as root that would read the file, drop root privs, and then start the webserver.

Usually Fabien gives a high level overview and dives into some of the pieces of the project that he finds interesting. I don't think this is the type of review that focuses on overall C practices unless it's something really unique to the project.

Here's a neat (and totally unpythonic) way to bring that down a bit:

    python -m timeit --setup "from collections import namedtuple; Point = namedtuple('Point', ['x', 'y']); p = Point(x=0, y=0); get_y = Point.y.fget; get_x = Point.x.fget" "get_x(p) + get_y(p)"

Hmmm, that looks oddly familiar: http://use.perl.org/use.perl.org/_Aristotle/journal/33448.ht...

I tried this once as well and I might have stuck with it if chrome had a quick "add to js whitelist" button. Instead you have to fiddle around in the settings menu which just takes way too long.

I ran across this yesterday (posted in 2011) about how google is using clang to automatically refactor their codebase: http://www.youtube.com/watch?v=mVbDzTM21BQ

I've accidentally pasted lines with --username --password into source files because of a sensitive scroll wheel... That's pretty embarrassing in code review.

What source are you looking for exactly? The piece was about the Dark Snow Project, and the journalist trekked around Greenland with the guy that founded it.

This is nonsense. You should judge the article on its own merit.

IIRC it's the way to switch between tabs

Swipe to navigate is a terrible design antipattern. I can always tell when I'm reading a blog on blogspot because it'll dump me to the adjacent article when I'm scrolling or trying to zoom in on an image. I dumped chrome on ios because they didn't get it right either.

Don't do that please.

This was a neat article about a hackathon, and it seems like you're ignoring the whole point of it to push a holy war against coffee. Also, you broke the several-month long streak during which I hadn't seen the word "brogrammer" used, so thanks for that as well.

That's really fantastic that he was able to save his work from 28 years ago. I really wish I had the foresight to do the same from only five years ago, though I expect the result would be less like unearthing buried treasure and more like finding what stinks in the fridge.

Do you mean "do as little with them as possible"? It sounded to me as if you're saying he's trying to make the least amount of dollars he can.

To make matters worse, after C&P rolled out some banks would just flat out refuse to reimburse cardholders for fraudulent charges. They claimed the system was bulletproof and if fraud did happen, then it was the cardholder's fault: http://en.wikipedia.org/wiki/Chip_and_PIN#Banks.27_liability

If you're using eclipse, I found this plugin helps a bit: http://marketplace.eclipse.org/content/java-source-attacher . For most popular jars you can just right click and hit "attach source".

It definitely depends on the language. The author mentions spring security so I'm assuming he's using java, where it's a massive PITA to manually hunt down and attach the sources to the 50 jars you have in lib/. Maven probably has something that will automatically do this but if your legacy project is just using ant then you're pretty much stuck.

I thought Dell submitted most of those patches to fix bugs in the kernel, not its own drivers.

I think the extra effort he means is that Dell had to hack in additional fixes for the trackpad, wifi, bluetooth and monitor brightness: https://launchpad.net/~canonical-hwe-team/+archive/sputnik-k... . They were all merged upstream, but if Dell had not done it, it would have been the customer spending hours on google or reading through threads on the ubuntu forums (I can speak from experience).

I read "ultrabook" as "any high-end laptop not made by apple".