reynolds

Varnish is heavily threaded. It maintains queues where connections are put and worker threads pull them out. It is expected that a single connection gets a dedicated thread.

When a page is requested, some of the links (fnid=X) map to functions on the server that are held in memory. The id (X) is a key that is associated with a continuation for a specific "session".

It's not a bug, but a feature of Arc and other continuation servers. If continuations aren't reclaimed by the server (and those links invalidated), the server runs out of RAM.

Expired links are the result of Arc's web server reclaiming stale continuations. It's a side effect of creating a web application using a continuation-based server.

I don't spend my vacations working for other people. You also can't expect a contractor to be passionate about your company when they have no vested interest in its success.

Fair enough.

I personally feel that Dropbox removing files from someone's account is completely wrong, regardless of your ToS. Your service is there to backup files. When you delete references to files from someone else's accounts, you're violating the trust that people put in your service.

I finally jumped into ios development. Not really on the side since it's been my main focus lately. I went from not knowing any ios stuff to launching my first app in a week. I'm about to put my third app on the app store. I'm not making much money but there's a ton of potential.

It was just bad wording on my part. What I tried to say was that there's a difference between giving $150k to a charity outright and donating money on behalf of the users who refer more users.

You're right. After jumping the gun I realized they're not "giving it to charity", but rather using it to acquire users.

It looks like $10/referral up to 100 referrals. If they make $20/user/mo this is a no brainer.

That makes sense. At least you guys are using it to get user referrals rather than throwing it in a wheel barrow and burning it :)

This seems like a slap in the face of the investors. Giving a chunk of your startup capital to a charity is basically throwing it away. Good for the charity but this startup seems too early stage to be giving that kind of cash away. I hope it pays off for them.

Felix is still working on Chicken. He's also active on the mailing list. There are a number of people actively working on it as well.

I love Chicken Scheme. The community is awesome and tightly knit. They're always looking to help out new users. I was new to the community awhile back and got a ton of solid feedback on some work I was doing in Chicken. I published epoll bindings for Chicken after figuring out how things were done in their community.

Moritz from the Chicken community recently published a Mongrel2 egg as well (http://wiki.call-cc.org/eggref/4/mongrel2)

It seems obvious that the person asking this question is a troll.

It's interesting to me that they turned down a $100M+ offer and raised at only a $25M pre-money valuation. Wouldn't it make more sense to raise at close to $100M pre-money?

I started with HyperCard as a kid. My parents got me a couple Java books but I never got into it. My first real programming experience was with Python when I was in high school. My dad told me that no one uses Python and that I should learn C++, so that's what I did. That was over 10 years ago so things have changed a bit with regard to Python :)

These days my favorites are C, Python, and Scheme.

There are a lot of reasons to believe we're in another startup bubble, but at least Cheezburger is profitable.

Private IPO sounds kind of funny to me. IPO literally means Initial Public Offering.

I wrote a blog post awhile back that was really controversial. It was about code ownership. People were split down the middle on hating code ownership or loving it. I think some people even took it to mean actual owning of the code.

I was coming from kind of a dictator view where I accept patches rather than blindingly letting people commit changes to projects i'm responsible for.

My mom knows what delicious is. That's pretty mainstream in my book.

I'm not sure that I agree with the claim that retraining someone for a short while can place that person at a similar level of mastery as his or her high-tech peers. It may work for entry level positions. And even then they may not have the same motivations as someone devoted to their craft.

I do a lot of Python and have even worked on my own web frameworks before. I've done web development with many different Python web frameworks. I also don't consider myself just a web developer.

I was recently interviewing for a Python/Django developer position and was instead offered a SDET position (with the possibility of becoming a developer) because I don't have experience with Django.

I turned them down because it felt like an insult.

If you're targeting a specific platform, it can make sense to use a virtual machine instead of installing that platform on your development machine. As far as code is concerned, it doesn't care that you're running it on a Linux VM inside a Windows dev machine rather than directly on a Linux dev machine.

That being said, if you're trying to target Mac or Windows, it's generally a good idea to use that specific platform's tools. It's reasonable to develop Windows and Mac apps on Linux but it may not be the best environment to do so. It's really up to you as a developer to determine the best toolchain and processes based on what you're trying to accomplish.

The way I do it is to write all of the code locally and use git/svn/hg/whatever locally. I then rsync my changes to the virtual machine when I'm ready to test.

It's actually something I've been working on that I want to put up on github. I hacked up a Python version of it as a proof-of-concept but rewrote it in C as an installable executable using autotools.

Basically I tell it which directory I want it to watch for changes and it does automatic syncing by piping rsync. I keep a local and remote signature of the files and their timestamps. When it first starts up, it pulls the server sig file and compares it to the local one. If there's a mismatch, the server is updated. From there it manages the signatures locally until they're different.

Writing the sig file to the remote server is done in the same rsync pass because it's stored in the local directory as a dotfile.

I realize rsync does its own checksums, but using my own crude signature files makes it so I don't have to keep calling rsync. I only call it when something changes.

I also have some stuff I'm working on that ties into auto-restarting servers when syncing finishes, rolling server deployments for no downtime, db migrations, etc.

I'm not sure if this helps in your situation, but when I'm running VMware I don't share any local directories. I use rsync to manage the virtual server directories since it's similar to how I deploy to my remote servers.

Having to manually "deploy" code to a local virtual server can be tedious, so running a background process that watches your local directories for changes helps a lot.

That nugget of useful information is also explained (in a lot more detail) in one of the first chapters of the "Introduction to Algorithms" book.

When I was working through the book (and I still am to some extent), it was basically beat into my brain that some algorithms with worse performance could perform better than faster algorithms for a small enough n.

These types of rants always strike me as being ignorant. It seems like it's just taking this one person's view of software development and trying to apply it to everything. To me, it's on the same level as saying "real programmers use X".

Achieve.ly. I had a few people interested in using it but I took a contracting position and had to stop working on it. I have a few people waiting to use it too. I'm probably just going to open it up publicly to let the handful of people use it.

I'm also working on some client/server stuff for automated web deployments.