Ruby on Rails
Ta-da List


May 03, 19:10

Backpack is now available

Backpack is out. Available. Here. Launched. Pushed. Online. Ya, know, yours for the playing.

Challenge by JB on May 03, 19:43

It was working ok, albeit slowly, for me in the beginning. Now all formatting has disappeared as well as images, so I get bleak, white pages.

Either Rails isn't as scalable as hoped, or the servers are getting hit TREMENDOUSLY hard.

Challenge by David Heinemeier Hansson on May 03, 21:10

We moved the static parts (images, stylesheets) off to another server and its on a non-standard port that's tricking some firewalls. We're fixing that now. Rails is scaling surprisingly well. The slowdown problems we had this morning was Apache being swamped. Once we got lighttpd into the picture, all was well.

Challenge by JB on May 03, 21:18

That's definitely good to hear. I wish you guys the best on this. I'd make sure to consider doing a "lessons learned" summary in the next few days to make sure folks who are experiencing the same things I am don't get the wrong impression....they may take it as a failure of RoR (like I initially did).

Keep up the good work!

Challenge by George on May 03, 21:35

Nice app, but you guys obviously need to get your head out of the sand and use your great skills to tackle something bigger. Look at who got bought out for $30+ million. Flickr and OddPost. They tackle things that everybody cares about -- photos and email -- but just do a better job than people before them. Look for something that is already a huge area (rather than trying to find an untapped space like personal organization) and just out compete people. That's success.

I'm impressed by the slickness of backpackit but it's something I'd never ever use. And I love wiki's, delicious and plenty of geeky things.

Challenge by PJ on May 03, 22:18

I actually just got through with an entry in my blog concerning just what the above poster mentioned. Sure, it's a fine app, but you guys can do much better and should aim higher the next time around.

Challenge by Matt L. on May 03, 22:41

Backpack is fantastic. 37signals has raised the bar yet again. I want to spread the word to some of the BB's I visit regularly on the web, but don't want to contribute to the already heavy traffic of the first day. Does 37signals prefer to hold off on Backpack evangelism until the initial rush dies down, or do they just say "bring it on!" ;)

Challenge by Lee on May 03, 22:48

Wow, I'm really really impressed with backpack. Can I come work for you guys! ;-) This is something I'd actually consider upgrading and paying for which is saying a lot for me.

I'm really curious about how you handle scalability on rails. IE are you load balancing apps like basecamp and backpack, session management (drb, sql, something else)? I have not been able to find a tremendous amount of information on it around the net. If someone that knows would send me some links I'd really appreciate it. I've been pushing rails development at my company for a while not and everyone seems slightly concerned about scalability. I've developed a few prototype apps that have drawn a lot of interest from other developers here and now we're trying to find out how far we can go with it before we need to look into load balancing etc.

Challenge by Sean on May 03, 22:51

David, I'm curious if you know of any articles (or are planning on writing one) regarding scaling Rails apps? I'm new to Rails, coming from a background in client-side software development and I have no idea how to set things up to handle these massive loads (obviously for development the DB, web server and rails are all running on the same machine). Are there good resources around for deploying Rails applications? I'm sure you've dealt with this for Backpack, Basecamp, etc.


Challenge by Sam on May 03, 23:21

Something similar to Backpack would be awesome for a wiki.

Challenge by Nigel Thorne on May 04, 0:21

Hi Guys,

First.. Love the site! I believe it is a really importnat step in how the internet will be used by individuals and communities in the future.

Second.. Backpack has the potential to be all things to all people.

* News Group : a page shared with specific users where they only add notes.

* Wiki: collection of shared pages

* Blog: a page of notes, publicly shared. (Missing is the ability to consume notes and emails on a public page as an RSS feed)

** It seems to me that notes and emails could be the samething? or emails could optionally be displayed as notes (possibly conditional on subject being "Note:..."

* Web Email : forward email to the page's email address, and its viewable from the web.

There is so much potential in your application. It's great to see.
Thanks for making it publically available.


Challenge by Randy on May 04, 1:20

Look at who got bought out for $30+ million. Flickr and OddPost.

Maybe 37s's goal isn't building to sell, but rather to build to build and be profitable on their own? If you think success is defined by how much you sell out for perhaps you should rework your definition. Maybe 37s can make $5+ mil a year on their own, on their own terms, and do it their way. To me that's true success.

Challenge by on May 04, 2:44

it is nice app but the haters have a point. Photos and email have a much larger base of potential customers to work with.

Everyone wants to share photos online but only a fraction of people on earth care about being organized. An even smaller fraction of people need to set up SMS alerts to remember to pick up lunch. :)

Making $5M a year on your own is very hard, even moreso when selling consumer priced services.

In order to gross $5M a year off of backpack, 37s would need 21,000 users on the $20/m plan. A very generous estimate would say that 5% of total users would be on the $20/m plan.

37s would need half a million users to net $5M a year. Once you have 500,000 users work starts to become a real headache so you pretty much want to sell out, anyway.

The last I read, basecamp had about 6600 paying customers.

Challenge by Mike on May 04, 3:39

Maybe they are not so much worried about a $30M payout as they are about helping folks deal with their world a little bit better.

Challenge by Jesse on May 04, 7:21

37signals has a history of continuing to improve their products after release, responding to customer feedback. I think we can expect some evolution to this core backpack application.

Also, it should be remembered that not everyone defines success in monetary units; those that do should be hesitant to criticize others who do not share their ideals.

Challenge by Morten Wittrock on May 04, 9:34

>Also, it should be remembered that not everyone defines
>success in monetary units

Hippie ;-)

Challenge by Morten on May 04, 10:22

I think a lot of people are curious on how to scale Rails applications. A whitepaper for doing so would be very welcome.

I imagine that eg. Basecamp can be scaled easily infrastructurally (if you have this sub-domain, we proxy you to this box). But for something like 43things, I'm curious on what load they're handling and how they do it.

If I were to give recommendations for a software project with a large budget, I need a clear picture of the capabilities of my choice of technology. Less than a handful of Rails sites have the experience to draw that picture, please do. The lighttpd/fastcgi method is well defined, the caching looks good, but Cameron Purdy does raise some valid pain points in his "FastCGI, not so fast" post - I still hope that someone can explain how to scale the database connections beyond a couple of hundred (this may even be a non-issue for MySQL).

Challenge by Amit on May 04, 12:46

Congratulation David- nice addition to your product family!
I am curious why you have used PHP for some of the pages (e.g. the example) vs 100% rails? Is it a limitation of Rails?

Challenge by Amit on May 04, 12:46

Congratulations David- nice addition to your product family!
I am curious why you have used PHP for some of the pages (e.g. the example) vs 100% rails? Is it a limitation of Rails?

Challenge by Thomas on May 04, 13:19

I'm sorry to say, but I don't really "get" backpack. Perhaps I just don't have the problem it is intended to solve.

Challenge by on May 04, 15:06

I don't get it either. Even if I did, it seems there are a couple major missing features: print view and search.

Challenge by JF on May 04, 17:14

Even if I did, it seems there are a couple major missing features: print view and search.

Just print any page with your browser. There are printer stylesheets in place so the page prints out nicely without any extra interface stuff.

Challenge by on May 04, 17:19

it is nice app but the haters have a point. Photos and email have a much larger base of potential customers to work with.

Seesh. What point is that? That the only thing ever worth doing online is building a photo sharing app or an email client? Wow, that's smart.

Challenge by rick on May 04, 22:08

From reading David's posts at least, Backpack is probably something that the Signals want to use personally. There isn't anything currently that hits the spot just right, so they wrote it themselves. If they can use this to free themselves from the "tyranny of client work" then more power to them.

Besides, there are enough photo sharing sites. I think Flickr's doing a good enough job on their own.

Challenge by Peter Cooper on June 03, 14:49

Am I missing something? Wasn't Backpack launched a few weeks ago?

And on the business side, they want the publicity rather than entirely money. There are things I do for this purpose too. The more profitable (even if they're not multi-million $ businesses) products they have, the more things they have to fall back on. If a competitor comes along and totally kills Backpack's market (not likely), then they still have Basecamp to fall back onto. Building up a suite of profitable, low-maintenance products and businesses is less risky than working on a single product, and means you have collateral to use to make riskier decisions in the future.

Challenge by Peter Cooper on June 03, 14:51

Ah, excuse my question above.. my newsreader showed this post as being new and being 'posted today'.. I came here, saw May 3, mistakenly thought it was May, and, well.. :)