Slashdot Mirror

I wanted to use xmbase_grok to create and manage a new simple database (which is what it was designed for). After fixing some bugs, I found that Motif (or at least the current version of openmotif, and, judging by earlier patches from other people, versions from years ago as well) cannot provide a proper multi-line text widget. For this reason alone, I ported xmbase_grok to Qt, which I actually don't like all that much, but at least it doesn't have broken multi-line text widgets.

Easy to follow instructions to deploy it:

https://bitbucket.org/cse-asse...

Reference manual in PDF:

https://bitbucket.org/cse-asse...

Easy to follow instructions to deploy it:

https://bitbucket.org/cse-asse...

Reference manual in PDF:

https://bitbucket.org/cse-asse...

'Execution'
The service uses our generic icap interface to send files to the proxy server for analysis and report the results back to the user.

So, does this mean I hack myself by the russians everytime I use it?

Here is the bits that matter, the source.
https://bitbucket.org/cse-asse...

Anyone can run their own XMPP server which will talk to any other XMPP service in a similar way to email

Any email server can refuse to accept connections from other email servers. One example of this is blacklisting known sources of unrequested advertisements. Likewise, any XMPP server can refuse to accept connections from other XMPP servers. Spam control is ostensibly why Google Talk defederated from other XMPP servers when it became Hangouts and switched to a proprietary protocol.

Not open per se, but..."source available" happens somewhat regularly. Ex: Star Wars Galaxies https://bitbucket.org/swgmaste... Others leak on the torrent sites and ragezone. Unfortunately they no longer get shared on betaarchive, as that recent BS story about "Windows Source Code" forced them to pull the source archives for things that were otherwise abandonware.

This is not surprising ! We discovered recently some "billion-laughs"-style DOS attacks that exploit vulnerabilities in Java, and ported some of them to .NET and Ruby. Details here: http://drops.dagstuhl.de/opus/... (paper, there is also an artefact to run attacks in a VM), and the source code is here: https://bitbucket.org/jensdiet... . We did have some problems porting this from Java to .NET but managed eventually. Interestingly, some of these problems were caused by a bug in .NET: a broken contract between equals and hashcode (see https://github.com/dotnet/core...) .

The purple-hangouts plugin works fine for my requirements.

Pidgin (or any other libpurple-based client, like Adium) with the purple-hangouts plugin works fine for my requirements.

If you need rebase in Subversion, there is a Python script that can do it https://bitbucket.org/x31eq/om...

Serious question, is there a latex-like tool for making presentations.

There are several of them. The most common is probably Beamer: https://bitbucket.org/rivanvx/...

A couple of others are mentioned on the Beamer Wikipedia page: https://en.wikipedia.org/wiki/...

shamilgumirov / presto-opera / Downloads — Bitbucket

The crucial, extremely illegal link: https://bitbucket.org/prestoco... And it compiles on modern linux, whoa. Can we have opensource, non-webkit, non-gecko browser now? Seriously, the slowness and resource of post-opera browsers is getting ridiculous - explorer is the fastest explorer at this point (not by js benchmarks, but overall user experience is far snappier).

Cameras already use processors that have nearly the feature set of a general-purpose CPU. (Canon's DIGIC is x86, and DIGIC II is ARM.) They run actual firmware. In fact, they often run an embedded operating system (e.g., VxWorks). That firmware can implement arbitrary features. Take as an example. You can see in the source code that it is not, in fact, simply enabling and disabling existing functions.

CPUs are slow to perform encryption because it's a lot of bit-level modification. CPUs don't have the instructions necessary to do those operations efficiently. That's why they added AES-specific instructions to x86. You can implement encryption in hardware, though, in which case it is very fast. High performance in hardware implementations is a major design feature -- it was required in DES, and it was required in AES.

False

In case you're up to fiddling with this project yourself, the source code (in a more manageable format usable with Scrolling Game Development Kit 2) is available at https://bitbucket.org/bluemonk.... I'm not sure I have the motivation to maintain it much any more myself. Unfortunately, Scrolling Game Development Kit 2 requires Windows, so you wouldn't be able to use that from OS X. I tried porting SGDK2 to other platforms (namely Linux) once long ago, but there was some difference in the way the Microsoft implementation of .NET treated serialized datasets that made me give up on that idea. I wonder if I would have more success with the new generations of .NET.

Business Insider didn't just become an authoritative source of news just because it is saying what Slashdot wants to hear.

- "Out with flat org structure based purely on meritocracy" Garbage. That's stuff of fantasies, no large commercial software company has a structure like that. Slashdot makes 'merit based hiring' sound like some kind of panacea - there is no such thing. Maybe GitHub are going the wrong way but this sort of description sounds like it came from someone with an axe to grind. No start up retains that cosy 'smart people you love to work with' feel.

- For the people saying just use a different hosting service, almost every worthwhile open source project is hosted on Github and tracks issues and releases on Github. ALL major companies use Github when they decide to go open source with a project. Guess where Apple put Swift? Microsoft when they wanted to develop an OpenSSH port publicly? Netflix? Yelp? Google's Tensor Flow?

- Alternatives - let's not even mention SourceForge. Bitbucket? Look at how terribly cluttered their UI is compared to Github: https://bitbucket.org/atlassia... And Github has massive first mover advantage here. I can't believe how awful Github's notification system is - I can't set up notifications to just keep track of new releases in a project for example.

- What some here hate is that GitHub is no longer focused on the traditional open source developer audience (if it ever was). 'Enterprise focused company' means what it says on the label. Yes they will have a massive sales force. Yes they are exploiting the brand name to sell an enterprise product that is way more expensive per user than their competitors (Bitbucket and Gitlab). But you know what - better than bundling fucking adware with downloads from their website.

- On the same point, they don't care much for the 'Git isn't server based, why do you need Github to host stuff' audience, or the 'you can take my eMacs from my cold dead hands'. They've put significant effort into their app, available on all platforms (yes an app - for 'developers' don't know how to run git clone or configure SSH keys. Snigger). You know what - they don't care. I heard from a friend recently how working in a major bank, their data science and modelling teams write code and don't use any source control. That's their target audience and that's where their sales people will make them money. I have lost count of the number of perfectly intelligent people I have dealt with who can't get their heads around Git or cannot be bothered to.

- Github doing what's best for Github, and when they do their sales pitch, a couple of slides of how Google hosts their projects on Github rather than the crappy code.google.com does not hurt. And I don't terribly care, compared to the products I have seen sales people sell successfully, Github is like vaccines - it's a good thing despite how it gets sold. A collaboration tool is a pretty damn good pitch.

- On the eMacs thing, Github released an Open Source plugin friendly editor called Atom. And I like it, I like it a lot. Github Page is pretty neat. Git LFS is awesome and works seamlessly for versioning large files and keeping them in the same repo - much better than the half baked git-annex option some projects used. It definitely does not look like they are out of ideas, despite apparently carrying a massive baggage of diversity based incompetent hires if Slashdot is to be believed.

- Look at this blog entry about a doctor who likes to code: https://github.com/blog/2103-m.... In commercial terms, you can't fault their choice of going for the much bigger market rather than sticking to trying to sell to 'pure' software / IT firms.

- Look at their blog, the huge list of integrations. They're not asleep at the wheel.

- Another one about their services team:

so wait, you are unhappy that we can setup our own OS on that thing? And to fix that, you are proposing to *restrict* the software you can run on it so that you can't modify it... that doesn't keep cisco routers from getting owned, or any other proprietary device from getting hacked, as far as i know.

there are litterally millions of home routers that run a "limited set of well documented functions" that are regularly abused for DDOS attacks to a complete port scan of the entire internet. and there are hundreds of people trying to fix those machines in various ways, either by reverse-engineering the hardware and installing free software on it or by just fixing the proprietary crap that's shipped with those. at least this machine starts on the right foot: it ships with free software and allows you to run your own.

any machine comes with its own foot shooting device, whether it is its openness or the false feeling of security that it's fine black box that will never fail and never need to be upgraded.

not understanding and not being able to fix a device isn't a advantage in security, i thought we agreed on that...

Ever heard that saying, "Emacs is a great OS but it lacks a decent text editor?" Well, thanks to evil-mode, now it has one.

I haven't seen much argument about evil-mode vs vanilla emacs. I think the consensus is pretty much, "it's all emacs." Emacs users generally customize their keybindings anyway - evil-mode is just a bit more extreme than most.

As for me, I like vim. I instinctively use vi keys when editing. However, I need custom scripts for some of the work I do, and I hate vimscript. Emacs gives me an editor that uses vi keys but lets me do customizations in elisp. It also has some other kick ass features I've come to like (org-mode is amazing). I still use vim for quick editing and sysadmin type stuff, but I write code and documents in emacs.

While Go and Swift are interesting incremental improvements, they are not taking into account what we learned about programming languages. In many ways, these two languages seem firmly stuck in the 1980s. For example, Go has no generics, and as far as I can tell, Swift still does not have the kind of true generic types I introduced in XL in 2000, i.e. the possibility to call "ordered" all types that have a less than, and then define functions with "ordered" instead of having to use <T> all over the place just like in C++ (and please, could we stop using angle brackets?)

More generally, there was a lot to be learned from more dynamic languages deriving from Lisp. Being able to treat code as data (homoiconicity) completely changes things. It means your language can be extended in itself, just like Lisp integrated object-oriented capabilities effortlessly. It means you can do metaprogramming, introspection, reflection, dynamic code generation, in a natural way rather than with specialised ad-hoc features. All things that Go or Swift spectacularly fail to do.

A real language redesign does not bring you incremental benefits, it should bring orders of magnitude on many tasks. I speak from experience. In XL, I can do complex arithmetic in 11 lines of code. What about Swift or Go? Ask yourself why Go can't offer complex arithmetic as a library package? Similarly, in Tao3D, I can do things HTML5 just can't, in a much less verbose, much higher-level language, and simple animations take 30 times less code than in JavaScript. The 30x factor tells me that I invented something new. Many others can demonstrate similar innovation.

I fail to see benefits of a similar order of magnitude with Swift or Go, and it annoys me. Companies like Apple and Google have the means, if only the financial ones, to make bigger things happen, in particular when smaller teams like ours already did a lot of investigative work.

for hd video, wifi is NO SUBSTITUTE for wired enet.

Bull*.

YouTube provides full 1080p videos at around 3.5 Mbps. If you rip and encode your own blu-ray at home, you'll be on the high end if you come in around 8 Mbps. Netflix serves up 1080p 3D videos that customers can watch on original Playstation 3 models. I routinely rip blu-rays and then compress/encode them at high quality settings for use via iTunes Home Sharing with my Apple TV. I never have to deal with waiting for more than a few seconds before playback starts. You can watch 1080p YouTube on an 802.11b laptop, those PS3s only have 802.11g, and my Apple TV only came with 802.11n. All of this is possible today and working just fine.

It sounds to me like you're blaming WiFi for a problem that lies elsewhere (e.g. misconfigured VLC buffering settings, wrong tools for the job, etc.), since plenty of us have working setups** that deliver HD video over WiFi without issue.

* Unless you're working with uncompressed video, of course, which consumes orders of magnitude more resources. If memory serves, a 1080p uncompressed video feed can be as high as 1.6 Gbps, so if you're dealing with that sort of stuff, then of course you should use wired instead of WiFi. But since we're talking about MKVs for an at-home setup rather than live video capture from an event, I figured we were dealing with compressed HD video. ;)

** Regarding my setup, I just have a typical 802.11n router and an Apple TV for hardware. On the software side, I use MakeMKV with DTS-HD to FLAC encoding enabled to rip my discs. I then use Don Melton's scripts (with the --big flag set) to crop and transcode them, followed up by a quick pass through Subler to add any of the metadata for use in iTunes. After that, I just add them to iTunes, enable Home Sharing in iTunes, and then log in on my Apple TV to gain access to my entire iTunes library. Super simple and works great, even though none of it is wired.

ImpulseTracker had its source released.

https://bitbucket.org/braindam...

These are some of the worst and most uninformative commit messages I've ever seen...

1) Why are there so many commits to achieve the same thing?
2) Any commit message that is only a single line other than "fix typo" is a bad commit message

Seriously, even some of the worst/most incompetent Android kangers have written better commit messages than the shitpile of LKM removals I'm seeing there.

There was an outcry when Google Labs closed because people actually used stuff that came from there. Mozilla Labs, on the other hand...

I'm still actively using a browser plugin that was the result of one of those Labs experiments -- Ubiquity -- and I didn't even know about Labs dying. It was one of their "failed" tests, but it was popular enough with some users that someone still maintains it despite its abandonment by Mozilla five years ago.

Unlike Google Labs, nobody noticed Mozilla Labs shutting down because it didn't instantly remove access to features, not because it had nothing useful.

I'm more concerned that the death of Labs means they're going to stop trying new and interesting things, or at least become far more conservative about trying new things. Down that road lies Internet Explorer 4 through 6 (or beyond).

How did this post get modded +5 Insightful?

made his white listing process extremely transparent and even allowed users to vote on it

Bullshit. Until late 2013, end users had ZERO official information, other than the whitelist URL and a vague hints. Two years of practical silence. Perhaps a good strategy, actually.

allowing you to completely turn off all of monetization functions with a single check box

That opt-out checkbox was still buggy when Edge was released nine months later. There were several versions of Plus where the checkbox literally did nothing.

kept his add-on fully open source and under permissive license that allows you to fork it

He does not have the authority to move to a non-free license himself. And really, there would be no point except to cause another, potentially more damaging scandal.

continues to work on developing the add-on

Doing, what, exactly? Adblocking itself is a fairly trivial concept; most of the work is keeping up with browser compatibility. Take a look at the change log. The only new features added since the fork are URL typo correction and "emergency alerts". Two completely inappropriate features that have no business being in an adblocker. Everything else has been UI tweaking and bug fixes. And guess what...

unlike those who forked it

You'll also have to tell me who offers an alternative that is actually developing the add on. Whoever forked Adblock Edge is certainly not that person. It's still stuck in previous version of Adblock Plus apparently, because whoever forked it couldn't even be bothered to update his fork by copy pasting code from new version.

... Adblock Edge is still up to date. There are commits from TODAY, in fact. Edge does not synchronize version numbers with Plus. That would be retarded.

If you are this paranoid, then how could you trust him in the past? Surely you had no way of knowing back then what his monetization scheme was

Not sure of the exact point you're trying to make. The code is open and can be reviewed in minutes. It isn't about "trust", it's about whether you want to support him in his endeavor to insert himself between you and advertisers. Personally, I think his strategy is a lose-lose proposition for both consumers and advertisers. That is, the only beneficiary is going to be his company.

So tell me. Why should I not trust him. Be specific, and try no to sink into the old "but turned the monetization options on by default when he implemented them (and I won't tell you that he actually warned you about it upon installation because that goes against the message I'm trying to deliver), that makes him completely untrustworthy" hyperbole.

Straw man, red herring, poisoning the well, whatever. You're still committing a fallacy, trying to circumvent an actual argument.

A lot of people had perfectly valid, negative initial reactions to this feature being made opt-out instead of opt-in. It is a legitimate ethical complaint. Studies show that *most* users *never* change the default settings. Setting the default behavior to circumvent the advertised purpose of the product is dirty before you even factor in monetization. But let's do that anyway. You really don't see anything wrong with the dominant adblocking provider having a sponsored whitelist enabled by default for tens of millions of people? Draw the parallels to malware and physical security.

What pushed me enough to bother switching to an alternative, though, was Wladimir's own arrogance. From my brief time on the AdBlock Plus forums, it became clear that he had no respect for us. Why should he? We are now the product, not the end users. It's a cliche because it's so depressingly common.

The latest major

BitBucket is still around. Like GitHub, it's actually fairly good, even the free stuff.

Don't know why they're using .org though. Definitely .com material.

Contains also offer security.
I've used it to run tests safely on student submitted code (server: https://bitbucket.org/gajop/au..., docker images: https://github.com/gajop/gradi... and https://github.com/gajop/gradi...).
It's done automatically for practice tests (for when students would submit their solutions online), so I don't even look at the source.
I know it's not guaranteed to offer 100% security as they could potentially break out of the container, but it takes care of most attempts or just mistakes (like accidental on the disk writing where they shouldn't).

And as far as their benefit over LXC, well it's really easy to setup: https://github.com/gajop/gradi... and use: https://bitbucket.org/gajop/au...

Contains also offer security.
I've used it to run tests safely on student submitted code (server: https://bitbucket.org/gajop/au..., docker images: https://github.com/gajop/gradi... and https://github.com/gajop/gradi...).
It's done automatically for practice tests (for when students would submit their solutions online), so I don't even look at the source.
I know it's not guaranteed to offer 100% security as they could potentially break out of the container, but it takes care of most attempts or just mistakes (like accidental on the disk writing where they shouldn't).

And as far as their benefit over LXC, well it's really easy to setup: https://github.com/gajop/gradi... and use: https://bitbucket.org/gajop/au...

I saw a talk on Ceph at LISA '13 and it seemed pretty cool, except that afterward I wasn't able to find any documentation specifically about making backups of Ceph object stores.

In general, I think backup infrastructure on Linux isn't great. I'm working to make that better, but generating interest in backup infrastructure is a hard sell. Shameless plug:
https://bitbucket.org/gordonme...

Several people did. There's a fitbitd Linux project: https://bitbucket.org/benallar... I tried it a couple of months ago - it works.

Mercurial is missing a few key features that make it less appropriate for solving some difficult collaboration problems, and some of those turn out to be important to people outside of Linux kernel development too. That's the technical part underneath what you're dismissing as a fanboy phenomenon.

Allowing octopus merges and making it easy to rewrite/rebase local branches are two such important features. The way those fit together into git's remote tracking branch concept is a particularly useful mental model for widely distributed development. And the way git (but not Mercurial) forces explicit branch naming and sharing gets people thinking in a way that usefully leads toward using these features for collaborative development. Mercurial vs Git; it’s all in the branches covers most of the important area usefully.

Basically, git includes some weird features that were built specifically for the style of distributed development done by the Linux kernel. But those turn out to be similarly powerful for other widespread collaborative software efforts too. These things aren't really appreciated by people until they've used git seriously for a while, and therefore they don't show up on most of these formal "which DVCS should we switch to" documents done by people still evaluating their options. That link I just referenced goes over how Google completely missed out on that in their comparison. You can't just compare the opinion of new users. The interesting thing to compare is what git is capable of on a project with some number of serious git users involved.

Once you've gotten used to things like git's remote tracking branches, good rebasing practices for code sharing, and malleable history when in tough spots, it's impossible to take Mercurial seriously for some of those jobs. Eventually Mercurial got some of the rebasing features, but by that point it had already lost quite a bit of the mind share war. They still are struggling with how the branch name model mismatches what some people want.

Oh yes, and another quote, apparently this parody is a hate crime:

Modern Leftist @ModernLeftist 14 Dec
@mgedmin @bitbucket http://bitbucket.org/ has servers in UK and the material is illegal. Report it here https://secure.met.police.uk/hatecrime_internet/

No, there are code samples. Here's my fav: https://bitbucket.org/FeministSoftwareFoundation/c-plus-equality/src/mistress/fizzbuzz.Xe

That's funny, I thought the remedy to unwanted speech is complaining a bunch and getting its hosting provider to shut it down.

If anyone is looking for an even simpler solution, I've package up my own project and open-sourced it. It's called pave, and has few dependencies, just PyYaml and fabric:
        https://bitbucket.org/mixmastamyk/pave
Would love to get more feedback.

Your problem isn't the storage, it's whatever you are doing locally that is the issue. I've got tens of thousands of files backed up with no issues, across several devices.

You didn't mention your OS. I'll assume you are running Linux because if you are running WIndows/MacOS you are missing a fundamental weakness already.

On Linux, use EncFS which also has a nice GUI manager via GEncfsM for those that prefer it.

Using EncFS means you don't have to upload entire files when you edit them, only the changes are synced. This is efficient, open-source, and works perfectly.

Once EncFS is working, pick any cloud storage you want and sync the encrypted folder(s). I do it with Dropbox + symlinks and it is flawless, no issues for years now.

Ah, sorry, link's dead, here's a copy: http://internetcensus2012.bitbucket.org/

PostgreSQL's biggest disadvantage over MariaDB is that it's not a drop-in replacement for MySQL.

Postgres has a blackhole engine that loses data in unpredictable ways now. But that only emulates parts of MyISAM.

Sorry, I should probably have added a disclaimer that I'm involved in the development of the library as my signature apparently doesn't make it obvious enough: I'm the project lead.

So far we've built about a dozen application with LibGeoDecomp, including porting a dozen large scientific codes towards it. You're right that porting a code usually involves debugging. But that's inevitable when parallelizing a previously sequential code anyway. We don't claim to do magic, we just have some cool tricks up our sleeves. And that's a Good Thing(tm). Because those who claim to cast magic usually disperse just b/s while clever tricks can save you weeks (months even) of work. Here is what you don't have to do if you use LibGeoDecomp:

• You don't have to write a proven (and correct) parallelization that scales to 1850000 (that's 1.8M) MPI processes.
• You don't have to devise your own domain decomposition and load balancing scheme.
• You don't have to write scalable parallel IO and application-level checkpoint/restart code.
• ...and so on and so on. A more complete list is here.

As said, parallelizing a sequential code will almost always involve some sort of debugging, no matter which tool you use. But the library also brings a couple of facilities to ease that transition: 1. you can first adopt the SerialSimulator which performs no parallelization at all, but allows you to check the data transfer and callbacks. 2. you can then transition to those parallelization which run on a single node only (e.g. the CacheBlockingSimulator or the CudaSimulator) to check that there are no race conditions before (3.) you finally more to large scale systems using e.g. the HiParSimulator (used for full system runs on JUQUEEN, an IBM BG/Q and ATM the fastest European machine) or the HpxSimulator (used for runs on TACC's Intel Xeon Phi equipped Stampede; BTW: it's built on HPX, a parallel runtime to C++). 4. Finally you can piggy-back the TestCell onto your model, which will use checksums to validate the data the library gives back to your code.

Sorry, I should probably have added a disclaimer that I'm involved in the development of the library as my signature apparently doesn't make it obvious enough: I'm the project lead.

So far we've built about a dozen application with LibGeoDecomp, including porting a dozen large scientific codes towards it. You're right that porting a code usually involves debugging. But that's inevitable when parallelizing a previously sequential code anyway. We don't claim to do magic, we just have some cool tricks up our sleeves. And that's a Good Thing(tm). Because those who claim to cast magic usually disperse just b/s while clever tricks can save you weeks (months even) of work. Here is what you don't have to do if you use LibGeoDecomp:

• You don't have to write a proven (and correct) parallelization that scales to 1850000 (that's 1.8M) MPI processes.
• You don't have to devise your own domain decomposition and load balancing scheme.
• You don't have to write scalable parallel IO and application-level checkpoint/restart code.
• ...and so on and so on. A more complete list is here.

As said, parallelizing a sequential code will almost always involve some sort of debugging, no matter which tool you use. But the library also brings a couple of facilities to ease that transition: 1. you can first adopt the SerialSimulator which performs no parallelization at all, but allows you to check the data transfer and callbacks. 2. you can then transition to those parallelization which run on a single node only (e.g. the CacheBlockingSimulator or the CudaSimulator) to check that there are no race conditions before (3.) you finally more to large scale systems using e.g. the HiParSimulator (used for full system runs on JUQUEEN, an IBM BG/Q and ATM the fastest European machine) or the HpxSimulator (used for runs on TACC's Intel Xeon Phi equipped Stampede; BTW: it's built on HPX, a parallel runtime to C++). 4. Finally you can piggy-back the TestCell onto your model, which will use checksums to validate the data the library gives back to your code.

Sorry, I should probably have added a disclaimer that I'm involved in the development of the library as my signature apparently doesn't make it obvious enough: I'm the project lead.

So far we've built about a dozen application with LibGeoDecomp, including porting a dozen large scientific codes towards it. You're right that porting a code usually involves debugging. But that's inevitable when parallelizing a previously sequential code anyway. We don't claim to do magic, we just have some cool tricks up our sleeves. And that's a Good Thing(tm). Because those who claim to cast magic usually disperse just b/s while clever tricks can save you weeks (months even) of work. Here is what you don't have to do if you use LibGeoDecomp:

• You don't have to write a proven (and correct) parallelization that scales to 1850000 (that's 1.8M) MPI processes.
• You don't have to devise your own domain decomposition and load balancing scheme.
• You don't have to write scalable parallel IO and application-level checkpoint/restart code.
• ...and so on and so on. A more complete list is here.

As said, parallelizing a sequential code will almost always involve some sort of debugging, no matter which tool you use. But the library also brings a couple of facilities to ease that transition: 1. you can first adopt the SerialSimulator which performs no parallelization at all, but allows you to check the data transfer and callbacks. 2. you can then transition to those parallelization which run on a single node only (e.g. the CacheBlockingSimulator or the CudaSimulator) to check that there are no race conditions before (3.) you finally more to large scale systems using e.g. the HiParSimulator (used for full system runs on JUQUEEN, an IBM BG/Q and ATM the fastest European machine) or the HpxSimulator (used for runs on TACC's Intel Xeon Phi equipped Stampede; BTW: it's built on HPX, a parallel runtime to C++). 4. Finally you can piggy-back the TestCell onto your model, which will use checksums to validate the data the library gives back to your code.

Ask and you shall receive :-). I have more information on that than you'd probably like to know. The back-end is actually quite similar to the PunkSPIDER project's back-end and uses all of the same principles, most of the same open software as its base, and even reuses some of the code (in fact, once it's done I'll probably make the back-end of web 3.0 a part of PunkSPIDER 2.0 - free and open source of course). So with that said here's info on how PunkSPIDER was built, which should give you a solid start to how we're building the web 3.0 back-end:

(1) A link to the talk at ShmooCon on PunkSPIDER which gives more info than you'd ever want to know about the back-end: http://www.hyperiongray.com/shmoocon
(2) If you're in a rush you can read some basic stuff about it here: http://www.hyperiongray.com/node/18
(3) If you really want to get into it you can download PunkSCAN (the PunkSPIDER back-end) on bitbucket and take a look: https://bitbucket.org/punkspider/punkscan

And last but not least, if you want to know even more feel free to contact Hyperion Gray at punkspider@hyperiongray.com or follow me (Alejandro) at @DotSlashPunk on Twitter. Oh and thanks for the feedback on the buzzy name, it's meant to be a little over the top, but we'll keep your comment in mind!

Alex

Hey, I keep my source code in BitBucket.

Nope

according to https://bitbucket.org/multicoreware/x265/wiki/Home there should be some sort of b-frames support: "With -b 1 you activate a hack which enables a "random access" GOP structure which uses B frames."

https://bitbucket.org/multicoreware/x265

The 5 user limit before paying is for private repositories only. There are public repositories (Example: https://bitbucket.org/tortoisehg/thg/overview ), although their homepage definitely doesn't do the best job explaining this.

Having long been a fan of minimalist window managers like Fluxbox, a friend recently recommended xmonad, which I promptly installed; and since doing so I have been amazed at how much time I can spend not managing windows.

There's apparently also osxmonad for the 'traitors,' but I haven't tried that out yet.

BitBucket has a better pricing model for what you are trying to do. Free for up to ten users, which includes unlimited private repos.

Anything public is probably better hosted at GitHub, as people are more used to using stuff from there - the pricing is not too bad, but is per-repository so if you have a bunch of small things it could start to get expensive.