adamlj

Caching and Vary headers can be tricky to get right

I can't access a couple of US sites (like github, linkedin etc.) from my Swedish ISP Bahnhof.

I find it interesting to compare this compiled JS to the tiny recursive Ray Tracer previously posted here on HN [0]. This Scala->JS compilation is 214k characters while the tiny Ray Tracer is at <1k characters. It should be noted that the tiny Ray Tracer is heavily optimized and minified.

[0] http://www.gabrielgambetta.com/tiny_raytracer.html

I'm not sure Google is actively punishing this strategy but the text doesn't look good at all in the SERP. The title of the page is displayed as "Ｕｎｉｃｏｄｅ － Ｆｕｌｌｗｉｄｔｈ － Ｚｅｉｃｈｅｎ" [0]

[0] https://google.com/search?q=site:linkstrasse.de

"This is an important lesson to learn: Mistakes can happen any time a piece of code is modified."

From http://www.daemonology.net/blog/2011-01-18-tarsnap-critical-... regarding the Tarsnap bug.

Solution: write more unit tests.

This could turn ugly really fast. If Russia attacks Ukraine the US and Britain would have to protect the borders of Ukraine, according to a treaty signed in 1994 [0].

If the US don't follow the treaty I guess we will know how much these treatys really are worth.

[0] http://www.dailymail.co.uk/news/article-2570335/Former-Briti...

I have been working with, and implented, both Stripe and Paymill. I agree with the article that the only time you would go with Paymill is simply if Stripe is not yet available (and in those cases you are probably better off with a local competitor).

Paymill is just a bad copy of Stripe in regards of technology, customer support, processes etc. One example of this is the ID verification process. Stripe has this automated and it takes minutes to complete. With Paymill we had to send copies of several different ID documents to different people and the process took weeks.

Does it support handling multiple sites? If not, do you have any plans for adding multi site feature?

Great offer I guess but I seriously had to shut down the page after looking at it for 30s..

I really need this update! Other than the good stuff already mentioned, The DecimalField finally supports a comma as the separator.

Nice idea. I think it's extra interesting since I remember reading a debate here on HN about "bribing" OSS developers.

My NL sites are live but NYC DC seems to be completely offline. This is bad, only last Friday they had some more downtime.

As of now there's no news at http://www.digitaloceanstatus.com/

Finally Enums!

The way things are going we will probably not have any Save buttons at all soon.

About a week ago I got to experience Digital Oceans very tight suspension policy first hand. What happened was that one of the accounts I manage was suspended. I had to go through a very long and detailed validation process before they understood that they had done wrong (they admitted to doing an error and apologized). But when my account was activated again my droplet had been destroyed..

After some more time they managed to resolve this and I'm again a happy DO user but I wish that they take a look at their policies. Just the fear of knowing that they can shut you down by mistake for a day or two is bad enough to not use them. They should have a policy where they at least call you and talk to you before they do anything.

Wow the Bouldering is amazing. So much fun and so dangerous!

I had a look at the source and I must say I really prefer how Bootstrap does it.

Compare:

  <div id="all" class="standard" intent in-width: in-container: in-orientation:>
  <section id="smallCode" class="clearFix standard" intent in-width: in-container: data-pattern="2">
  <div id="pseudosmalltablet" intent in-pseudosmalltablet-class="toggleOrientation"  in-base-class="null"></div>

etc. to

  <div class="col-md-6">
  <div class="col-xs-6">
  <div class="hide-xs">

It sounds very likely that it was Ruby that consumed so much of the CPU.

By your logic people shouldn't be using Linux, Windows or any other software because people have found exploitable bugs in them.

There has never been a backdoor in Pidgin. Instead Pidgin has a very healthy security community and they release several CVE's each year.

I would use Pidgin over Tox without a blink since Pidgin has been around the block longer (almost 10 years now) while Tox hasn't released a single Security advisory yet. So I'm sure there's a lot of "low hanging fruit" exploits waiting to be found in Tox.

I made that switch too and Im really happy I did. Same feeling as switching from Wintendo to Linux. And gnome has gotten really good!

This was very nicely written and resonates with my thoughts. Get off the hamster wheel!

If you have to store CC data inhouse I would suggest storing it on a completly sepatate machine which only stores and charges cards. The only communication allowed from this box would then be Store this card, Charge the card with this token etc.

Seeing the hits/sec on the count down is pretty impressive

I always think of the Simon & Garfunkel song when hearing his name.

The real problem here is that this is extremely asymmetric. All the bad guys need to do is to find a single mistake from an employee, subcontractor, vendor etc.

Cool stuff! Now that everyone and their grandmother are building apps it makes sense to start selling shovels (as in selling shovels in a gold rush).

Fukushima disaster has not raised cancer risks [1]

We need more nuclear power and less nuclear scaremongering.

[1] http://www.rsc.org/chemistryworld/2013/06/un-report-fukushim...

It's really sad to see how effective they have been in ruining this society.

More like this: http://www.foreignpolicy.com/articles/2010/05/27/once_upon_a...

At least this is true right now but the above could easily turn out to be one of those "None will ever use more than X bytes of ram" quotes.

We are always talking about pivoting and fail fast so why get upset when Yahoo! starts practicing what we are preaching.