HNHacker News
TopNewBestAskShowJobs

mrshoe

4,866 karma · joined October 7, 2008

busy busy busy
submissionscomments
mrshoe··on GitHub's Unicorn Setup
Sure, now that HN has decided that enough time has elapsed to allow me to reply. :-)

Actually, antonovka did an excellent job explaining it above. The most important aspect is that threads are "lighter weight" than processes. They use less memory and are quicker to context switch (usually). The result of this lighter weight is that you can spawn more threads than you could processes, on the same hardware. And when you're using one thread/process per connection, that means more concurrent connections on the same hardware. So, if Unicorn used its exact same architecture, but replaced the worker processes with worker threads, you could scale much better.

Secondly, haproxy is written in C, which generally means it's going to perform much better and use a lot less memory than a Ruby webserver. This translates, once again, to more output from the same amount of hardware.

That's why I was pretty surprised to see that github would choose both Ruby and pre-fork over C and threads (or in the case of haproxy, async, which scales even better).

mrshoe··on GitHub's Unicorn Setup
Where were all you people in this thread the other day when I seemed to be the only one who understood the other side of this debate?!? :-P

I'm also very curious about the 16 workers per box model. Frankly, that github would ditch haproxy in favor of a pre-fork Ruby web server is shocking.

I'm just glad we have CherryPy in the Python world.

mrshoe··on Real-Time Collaboration Has Stalled...For Now
It's quite a bit different from group chat in Gtalk. The biggest difference is that the rooms are persistent and any time you log in you can scroll back to catch up on missed conversations. With Gtalk (or any IM service) group chats are usually ad-hoc and treated more like meetings: you start one, talk for a while, then close it. The idea here is to have a constant stream and have topic- or team-oriented rooms where people can always go to ask questions or get ideas from their coworkers.

It is, however, very similar to Campfire. Right now, the advantages over Campfire include performance (which is actually quite key for real-time conversations like this), and the ability to sit in multiple rooms at once and receive notifications about messages to you in background rooms (also pretty important, we've found). That said, we're in beta! So, we definitely plan on differentiating ourselves more in the coming months.

mrshoe··on Real-Time Collaboration Has Stalled...For Now
As all of us chatters and instant messengers grow up and enter the "real world", we are realizing that those forms of communication and collaboration can be really useful at work.

Unfortunately, the usual reaction is to get everyone in your office on yahoo/aim/gchat/msn messenger, which is far from ideal.

I think this space is bound to see quite a bit of growth in the near future, and it's far from a solved problem. That's why we're building http://shoptalkapp.com :)

mrshoe··on Minimizing Complexity In User Interfaces
This is a great article and it covers almost all the bases. The only thing I'd add is a discussion of UI refactoring. Over time, creeping featurism tends to take over. Your users will request every possible feature and your marketing department will want to add more check marks to your product's column in the feature matrix.

As this article points out, you have to be the bad guy and Just Say No most of the time. However, when a large subset of your users request a certain feature, the best response is often to refactor your design. Don't just add the feature. Figure out why they're requesting it. In other words, figure out why the current design is lacking. At that point you can come up with a new design which lets your users accomplish what they're trying to do.

User interfaces age like code bases in this way. If you keep adding to them, they get really ugly really quickly. If you instead take the time to refactor them, they can remain beautiful while becoming more capable.

mrshoe··on I like Unicorn (Rack HTTP server) because it's Unix
Oh, you were referring to the syscalls/portability tradeoff. In that case it's still not really a counter example, because Apache made the same decision in its early versions (Unix only). And that's a reasonable decision to make given the tradeoffs, I think.

The point that I was more upset with was the "threads are dead, pre-fork is the way to go" section of the article.

However, thinking about it more now, even the syscalls vs. portability arguments presented here are another example of forgetting the past (or just never being aware of it). After all the work from the Python devs to encapsulate syscalls in the standard library and provide a portable API to them, Jacob Kaplan-Moss says, "I’m a bit dismayed to see [syscalls] relegated to the dusty corners of our shiny dynamic languages."

mrshoe··on I like Unicorn (Rack HTTP server) because it's Unix
That's not really a counter example. nginx does indeed fork off child processes, but it uses async I/O in each child process. It doesn't use the traditional pre-fork process-per-connection model that Apache's pre-fork MPM uses or that Unicorn is using.
mrshoe··on I like Unicorn (Rack HTTP server) because it's Unix
There's a good reason most Ruby and Python projects don't rely heavily on system calls: portability. If you care at all about portability, it's just easier to not hit the system calls directly, otherwise you'll have to detect host OS and make sure you're using the system calls properly.

If threads are "out", then pre-fork is way out. Just look at the history of the Apache project. I realize this all happened before the RoR era, but Apache used to use a pre-fork MPM almost exclusively. In more recent years it has added the threaded MPM and the async MPM. They have also put in the work I mentioned above to achieve cross-platform compatibility.

I'm just using Apache as an example here; I'm not suggesting that we should all use Apache. It's just funny to me to see a post that basically says, "All the stuff we've been doing for the last 5 years is out. We should be doing the same stuff they were doing 15 years ago, but in Ruby this time around instead of C."

Maybe software trends are like music trends? Everything from 5 years ago is lame, but the stuff from 20 years ago is super groovy, man.

mrshoe··on Email Newsletters Are Serious Business
You don't have to run an email newsletter to make money in this business, either. Check out this very cool (and successful) piece of Mac software by HN user jhammer:

http://ethreesoftware.com/directmail/index.php

It does pretty much everything you'd need to run one of these newsletters.

mrshoe··on What "Super Achievers" Know About Time Management
Maybe there are some HN readers who are part of small teams trying to build profitable businesses from scratch and they need to figure out how to wear 10 different hats and get 100 hours of work done every 24 hours and still have time to sleep.
mrshoe··on What "Super Achievers" Know About Time Management
To-do lists can work for some people, but here's why they're dangerous:

1) The urgency-importance matrix theory (http://en.wikipedia.org/wiki/File:MerrillCoveyMatrix.png). Given the 2x2 matrix of (important, unimportant) x (urgent, not urgent), the two urgent quadrants tend to get done and the other 2 don't. In the case of the (unimportant, not urgent) quadrant, that's probably fine. But for the (important, not urgent) quadrant, this is a problem. To-do lists can make it even easier to focus on the urgent, short-term tasks while starving the important, long-term ones.

2) To-do lists provide fuel for the procrastination fire by helping you find ways to occupy yourself while you put off a big task that you don't want to do. See http://www.structuredprocrastination.com/

I think that's the crux of his argument. People who achieve a lot can see the big picture; they can see the long term. They don't let the small tasks get in their way.

mrshoe··on Unit testing in Coders at Work
The most important lesson to learn here is that all code, including tests, is a means to an end. A lot of people geek out about new programming languages, object oriented design, automated testing, TDD, XP, and various other methodologies du jour. If those things help you achieve your goal, that's great; but it's important to remember that they're (usually) not your goal in and of themselves.

I don't think jwz would say that unit testing is a bad idea. What he was trying to say when he dismissed them was that they were focused on making a browser, not on making a nice piece of code.

Similarly, Norvig was focused on solving sudokus, not on exploring methodologies that might be used to write the code which solves sudokus.

mrshoe··on Mark Zuckerberg: The evolution of a remarkable CEO
I would even sum up his advice as, "Do exactly the opposite of what these other smart people like Mitch Kapor have been telling you to do today."

Of course, as you mentioned, he arrived just in time for his talk, so he didn't even know that he was directly contradicting the advice of previous speakers.

He also scoffed at any form of testing when an audience member asked.

I was beyond unimpressed. Hopefully he redeems himself this year.

mrshoe··on Ask HN - What's your best startup idea?
http://familysearch.org

and in beta:

http://new.familysearch.org

http://pilot.familysearch.org

mrshoe··on On Being Narrow
In March I was shopping around for a new job. It was common to be asked why I was changing jobs (especially in this economy...). A big part of my answer was that I wanted to try something new. I wanted to make sure that my resume didn't pigeonhole me as a Python web developer. Even though I had been in that field for my first 4 years out of college, I had worked in very different domains previously and I wanted to keep my options open in the future. I enjoy Python web development, but to avoid being too narrow I decided that my next job would be with a different language in a different domain.

Most people understood that desire. A recruiter at one company where I interviewed disagreed with the strategy, though. She said it's ok to play around in a few different domains when you're young, but at some point you need to pick one and stick with it. That way you can be seen as expert in that domain and you can be considered eligible for higher level roles with more decision-making responsibility.

I suppose I can see her point, but I sincerely hope she's wrong. I want my horizons to always be broad.

mrshoe··on Geeks Try Google Wave, Have Mixed Feelings
The problem is definitely there, but, after using wave for a couple of months now, I don't think it's the right solution.

I would echo a lot of other reviews: it's too complex and ambitious. This applies to the protocol/platform as well as their initial client. There's just so much going on and it's trying to do so much at once that it's very unwieldy.

I think simplicity is an important feature if you want new software to gain traction, and especially if you want new protocols to be adopted. I think simpler tools like http://shoptalkapp.com (disclosure: co-founder) do a better job of solving the problem you describe. And the tools we use to build it, which are gradually being open sourced at http://dieselweb.org, make real-time web apps pretty accessible to developers.

Having used the wave client and read about the protocol, I just think both are too complex to really gain traction. All the bugs in Gooogle's own wave client are evidence that it's not an easy platform to write apps on, at least at this point. I'm sure they'll continue to pour tons of resources into it, and maybe they'll turn it around. Time will tell.

mrshoe··on Google Wave crashes on beach of overhype
First of all, the guys at Google have been padding this landing pretty well. At least for people who are paying attention. They're quick to stress that wave is a protocol and a platform with take-over-the-world potential and that this initial client that we're playing with is just one example of what can be done with it.

That said, I agree with this article completely. The wave client is an absolute nightmare to use. Stuff is blinking and flashing all over the place. People are adding comments up and down waves will-nilly and there's no good way to tell where changes are happening. There's a decent amount of latency with every single character I type, which basically kills my thought process and typing ability. Switching waves is a multi-second proposition that beach-balls FF on my Mac Pro.

Even if all these bugs were ironed out, I don't see why or how I would ever use it. I would have given pretty much exactly the same review a couple of months ago when I started using the wave sandbox, so whatever progress they've made hasn't made much of a difference so far. Maybe they're onto something and wave as a protocol will end up replacing a bunch of old web protocols which people have tried and failed to replace for years, but, at this point, I wouldn't bet on it.

mrshoe··on The 2009 TechCrunch Web App Survey Results
Looks like a really small sample size, about half of which are bootstrapped startups with zero users. I'm not sure how useful this data is.
mrshoe··on Is 37 Signals Selling Behavioral Targeting Data?
I honestly can't tell. The confusing nature of the post leads me to believe it's FUD, especially since the call to action at the end encourages me to use their product to protect me from the supposed evil.
mrshoe··on Ruby on Rails Workshop for Women
This reminds me of all the clever kids in school during Black History Month who would ask, "And when is White History Month?"

The canned answer is, "The other 11 months of the year are White History Months."

This answer has become trite at this point, but the argument at its core remains valid. Unfortunately, there are serious biases in our society. Correcting those biases is a worthy goal, and one we should pursue. In the mean time, we can use Black History Month and Ruby Workshops for Women to compensate for the biases and at least try to treat the symptoms.

mrshoe··on College Without High School
> My book is actually written for just those teens: the ones who have attempted to diligently follow school’s path but have only met disillusionment and boredom in the process. (Middle school is usually when this sets in.)

That was me. I firmly believe that middle school and high school did more damage than good for me, educationally. The only thing I really learned was how to get the A grade while doing as little work as possible. I could have learned and accomplished so much more in that time, and been better prepared for college.

However, I have a hard time asserting that I would have been better off skipping high school, overall. Education isn't the only thing you get out of high school. You learn how to handle yourself in a variety of social situations and with a variety of other people. High school might have handicapped my education, but skipping it would have handicapped my social skills.

mrshoe··on Leaving my job & starting my own thing
To chime in from the ShopTalk team, I would echo a lot of what axod has said. As a Campfire user, I'm sure you understand how important persistent group chat is to your organization. Unfortunately, most people aren't like you (yet)! We think there is a lot of room in this space, because more and more companies will realize how useful apps like ours are over the next few years. It's a relatively young, growing market.

Aside from that, there are a lot of improvements yet to be made in order to polish the group chat experience. One thing that we do, for example, is to allow users to sit in multiple rooms and be notified immediately about activity in background rooms. If you have more than one team using your account, that quickly becomes very important. That's just an example of one cool feature we've already implemented. We're still in beta, so there's a ton yet to come from us.

Thanks for checking it out.

mrshoe··on Leaving my job & starting my own thing
We're on it!
mrshoe··on Leaving my job & starting my own thing
We agree, and we think http://shoptalkapp.com will be the one to do it. ;)
mrshoe··on JWZ Unable to Distribute Two Free Pre Apps
jwz says this in the comments:

An important point here is that Palm-the-company used to not be stupid in this way. The way this worked on PalmOS was completely sane. They went and fucked up their whole process when they switched to WebOS.

I assume this is because they hired the majority of the WebOS management team from Apple, pushing out the people who did it right in the PalmOS years, and those new people are idiots.

I'm not sure it's safe to assume that this process is stupid. Palm did, indeed, copy all of this straight from Apple's playbook. Apple's App Store is wildly successful. It may have issues from the perspective of some developers, but that's not the only perspective that matters. Palm isn't trying to make a bunch of developers happy. They're trying to sell phones and make users happy.

Whether you can make users just as happy without requiring developers to jump through hoops remains to be seen. One could easily argue that it's possible, but I'd like to see evidence of that in the real world. I don't think the software ecosystems for desktop computers are a fair comparison. The ecosystems for older Palm devices, which jwz references here, were never nearly as successful as Apple's App Store. Maybe that suggests that the hoops somehow help the user experience? I don't know. But it looks like Palm is assuming that it does.

I agree that this whole process seems completely broken from jwz's perspective, but it's hard to say that Palm had people "who did it right in the PalmOS years" when they never got it right in the category that matters: number of app sales. That could be 100% attributable to the iPhone's superiority, not the App Store's, though.

mrshoe··on Here's an idea for Facebook. Get rid of subject lines in messages
The point of messages is that you can't see them, unless they're to/from you.
mrshoe··on jQuery Desktop
I want a desktop OS that runs web apps so well that it obviates the need for a browser, not an app that runs in the browser that emulates my current desktop OS.
mrshoe··on Ask HN: What do you do on weekend?
http://dieselweb.org/

http://shoptalkapp.com/

Ah, the joys of bootstrapping a startup on the side...

mrshoe··on Apple company culture in one license plate...
sama is at Infinite Loop right now? Cue the speculation about Apple buying Loopt or Loopt receiving special treatment on the iPhone.
mrshoe··on Efficient Editing With vim
Nooooo all of the secrets of our vim cult revealed in one easy-to-read document! Vim is hard, people! Trust me. You're not smart enough to use it.

:-P

← PreviousPage 9 of 14Next →