Ok, Rails has been out for some time now. I''ve kept an eye on it pretty closely. I''ve heard Basecamp, Robot Coop, 43 Things, etc. are running on it. However, those have been the same apps people point to when the dreaded scalability question comes up. Are there any new web apps (large scale) in Rails or is it still the same old one? Something like on the scale of Digg.com, Slashdot, Amazon, etc. With all the press Rails gets I was expecting to see a lot more Rails apps out there other than the 40+ todo lists and other blog type web apps. Please, I''m not trying to flame - its just that these are the same web apps I showed my boss 6 months ago and I certainly don''t want to go back and show the same ones. Surely there are more large scale RoR web apps out there by now? I see PHP powering a ton of big sites, I was kinda expecting RoRs to have a little more by now. -- Posted via http://www.ruby-forum.com/.
Ken McMyre wrote:> Ok, Rails has been out for some time now.I guess the idea is that Rails is the killer app for Ruby and now you want to know where''s the killer app for Rails? (Sorry, just trying to sum it up.) Killer app meaning one that causes ooh and aahs beyond the web developer mindset. I''m not sure there is one yet. Rails is just a tool. Rails went 1.0 at the end of last year and it has been undergoing some pretty big changes since its big release. It took nearly ten years from Ruby to Rails. Perhaps we need to give people time to figure out where all of this power can take us. We also need to see the javascript libraries become more powerful -- so the applications feel just like desktop applications. My guess is that we need about a year or so to see a killer app. This will give the interface conventions some time to settle, and the developers some time to roll out a slew of sites. A couple of hundred decent Rails sites could have a killer app in the bunch. -- Posted via http://www.ruby-forum.com/.
> Please, I''m not trying to flame - its just that these are the same web > apps I showed my boss 6 months ago and I certainly don''t want to go back > and show the same ones. Surely there are more large scale RoR web apps > out there by now? I see PHP powering a ton of big sites, I was kinda > expecting RoRs to have a little more by now. >I''m assuming that you''ve discussed all the ''rails scales'' arguments with your boss - how the suggested rails scaling methods are those methods already proven by yahoo and google etc etc etc? As for examples, how about http://www.rapidreporting.com/? According to the agile book (1st.ed), their systems are built to handle 2million+ mortgage applications per month. Also, you may have shown your boss those apps 6months ago, but all of those examples have increased their traffic dramatically in that time. I re-re-re-iterate what many great men have said before: hardware is cheap, developer time is not. The choice is about where you want to put your money and whether or not you''re willing to take a risk. I had the same choice a while back. I chose Rails and I''ve not looked back. Steve (steve has left the building :0) )
On 7/17/06, Stephen Bartholomew <steve@curve21.com> wrote:> > Please, I''m not trying to flame - its just that these are the same web > > apps I showed my boss 6 months ago and I certainly don''t want to go back > > and show the same ones. Surely there are more large scale RoR web apps > > out there by now? I see PHP powering a ton of big sites, I was kinda > > expecting RoRs to have a little more by now. > > > I''m assuming that you''ve discussed all the ''rails scales'' arguments with > your boss - how the suggested rails scaling methods are those methods > already proven by yahoo and google etc etc etc? > > As for examples, how about http://www.rapidreporting.com/? According to > the agile book (1st.ed), their systems are built to handle 2million+ > mortgage applications per month. > > Also, you may have shown your boss those apps 6months ago, but all of > those examples have increased their traffic dramatically in that time. > > I re-re-re-iterate what many great men have said before: hardware is > cheap, developer time is not. The choice is about where you want to put > your money and whether or not you''re willing to take a risk. I had the > same choice a while back. I chose Rails and I''ve not looked back. > > SteveDHH on the matter: http://article.gmane.org/gmane.comp.lang.ruby.rails/22562 Rails scales well, Rails with caching scales unbelievably. Steve is right, it''s easy to throw servers at a problem but as Brooks points out on the Mythical Man-Month you cannot throw programmers at a problem. People don''t scale well at all, so even a little scalability on a machine''s part is a welcome bonus. Show your boss shopify though, very cool stuff in my opinion and fairly new; I think it''s within the last 6 months. Cheers, Chuck Vose
> developers some time to roll out a slew of sites. A couple of hundred > decent Rails sites could have a killer app in the bunch.Hmm.. i don''t know whether it''s really about ''killer apps''. The true bedding of rails will be when developers take it up to create everyday sites. I run a web company and we make fairly boring cms/commerce sites for companies that would never be considered ''killer'' - even if they do have high traffic. We use rails makes us which in turn makes our clients happy. More people will discover this and it''s then that we''ll start seeing some high-traffic, ordinary sites - like the rapid reporting guys that i mentioned in my last post. It''s never going to come top of the ''Web 2.0 uber-site'' list, but it does give a good everyday example of a rails app performing well. That''s not to say i''m not into killer web apps of course - i''m just jealous that i haven''t got one :0) Steve
On Jul 17, 2006, at 12:46 PM, Ken McMyre wrote:> Please, I''m not trying to flame - its just that these are the same web > apps I showed my boss 6 months ago and I certainly don''t want to go > back > and show the same ones. Surely there are more large scale RoR web > apps > out there by now? I see PHP powering a ton of big sites, I was kinda > expecting RoRs to have a little more by now.You''re not going about this the right way. Rails works the way you make it work (for the most part) and it relies on multiple technologies (LAMP clustering primarily) that are tried and true in the large scale world. There''s nothing to discuss, nothing to prove, other than that the Rails community can produce some apps that MANY people will use. But that''s not your concern, right? As DHH put it at one point (paraphrased): There''s nothing worth talking about with respect to Rails scaling, because it''s BORING. You need to remember that Yahoo and SlashDot are amazing because: 1) They''re REALLY BIG. 2) They were REALLY BIG in the 90''s, when computers were like dinosaurs compared to the current systems we have. Nobody I''ve talked to with scaling experience has any particular concerns about Rails'' scalabillity. Yes, there are some areas that could be improved, but they''re just matters efficiency, not scalability. -- -- Tom Mornini
It''s not about where are the BIG web apps? PHP has been around MUCH longer than rails, as a result many of the new web ideas were done in PHP, because that''s when the web was really taking off. My guess would be that if rails was around at that time they would be done in rails. I''ve programmed in PHP for over 7 years. I switched to rails about 6 months ago and will never look back at PHP. In the past 6 months my company and I have made 2 programs in rails that are making us well over $30K a month. Doing those apps in PHP would have taken us a good 8 - 10 months each. Take a look at basecamp. David said he did that in 2 months. Comparing PHP to rails in terms of big web apps is an unfair comparison because PHP has been around for so much longer. Thank You, Ben Johnson E: bjohnson@contuitive.com On Jul 17, 2006, at 3:22 PM, Tom Mornini wrote:> On Jul 17, 2006, at 12:46 PM, Ken McMyre wrote: > >> Please, I''m not trying to flame - its just that these are the same >> web >> apps I showed my boss 6 months ago and I certainly don''t want to >> go back >> and show the same ones. Surely there are more large scale RoR web >> apps >> out there by now? I see PHP powering a ton of big sites, I was kinda >> expecting RoRs to have a little more by now. > > You''re not going about this the right way. > > Rails works the way you make it work (for the most part) > and it relies on multiple technologies (LAMP clustering > primarily) that are tried and true in the large scale > world. > > There''s nothing to discuss, nothing to prove, other than > that the Rails community can produce some apps that MANY > people will use. But that''s not your concern, right? > > As DHH put it at one point (paraphrased): There''s nothing > worth talking about with respect to Rails scaling, because > it''s BORING. > > You need to remember that Yahoo and SlashDot are amazing > because: > > 1) They''re REALLY BIG. > 2) They were REALLY BIG in the 90''s, when computers were > like dinosaurs compared to the current systems we have. > > Nobody I''ve talked to with scaling experience has any > particular concerns about Rails'' scalabillity. Yes, there > are some areas that could be improved, but they''re just > matters efficiency, not scalability. > > -- > -- Tom Mornini > > _______________________________________________ > Rails mailing list > Rails@lists.rubyonrails.org > http://lists.rubyonrails.org/mailman/listinfo/rails-------------- next part -------------- An HTML attachment was scrubbed... URL: http://wrath.rubyonrails.org/pipermail/rails/attachments/20060717/7976e193/attachment.html
> Also, you may have shown your boss those apps 6months ago, but all of > those examples have increased their traffic dramatically in that time.Besides the 37signals apps I''ve never seen such a list ... can anyone offer up some examples? We''re a rails app and OUR CUSTOMERS frequently ask us for examples of other successful apps using RoR. Not surprisingly, they''re looking for reassurance that this cutting edge tool will meet their needs in all facets, not just performance and scalability (e.g. reliability, security, etc). A list of sites would be most helpful because they''re not interested in hearing a technical breakdown (e.g. requests per/xyz, jargon, jargon); they want real world examples. Thanks in advance.
Isn''t Shoppify a fairly huge web app? Or at least potentially so? -- Posted via http://www.ruby-forum.com/.
Jon Gretar Borgthorsson
2006-Jul-18 01:33 UTC
[Rails] Re: Rails - where are the BIG web apps?
A good list is here.. http://wiki.rubyonrails.com/rails/pages/RealWorldUsage Among the big names are odeo.com, penny-arcade.com, calendarhub.com, writertopia.com, streeteasy.com and so on. Most of these have traffic that would make the average sysadmin blush. However if you are looking for names that everyone knows. Like ibm.com or microsoft.com. Then no. There are no names like that. But those sites arent exactly using cakePHP. All sites like that spend millions on their site and can afford making their site in assembly if the feel like it. On 7/17/06, Brittain <scott_brittain@hotmail.com> wrote:> > Also, you may have shown your boss those apps 6months ago, but all of > > those examples have increased their traffic dramatically in that time. > > Besides the 37signals apps I''ve never seen such a list ... can anyone offer up > some examples? We''re a rails app and OUR CUSTOMERS frequently ask us for > examples of other successful apps using RoR. > > Not surprisingly, they''re looking for reassurance that this cutting edge tool > will meet their needs in all facets, not just performance and scalability (e.g. > reliability, security, etc). > > A list of sites would be most helpful because they''re not interested in hearing > a technical breakdown (e.g. requests per/xyz, jargon, jargon); they want real > world examples. > > Thanks in advance. > > > > > _______________________________________________ > Rails mailing list > Rails@lists.rubyonrails.org > http://lists.rubyonrails.org/mailman/listinfo/rails >-- -------------- Jon Gretar Borgthorsson http://www.jongretar.net/
> Besides the 37signals apps I''ve never seen such a list ... can anyone offer up > some examples? We''re a rails app and OUR CUSTOMERS frequently ask us for > examples of other successful apps using RoR.in terms of VC-funded bay-area music-related web2.0 things, theres Odeo and Mog, and probably 50% of the sites on techcrunch that arent using JAVA or PHP are using rails. as far as enterprisey, rick bradley has talked about the conversion of a large app to rails: http://rewrite.rickbradley.com/pages/moving_to_rails, and presumably ThoughtWorks (and others) are using rails in enterprise situations. as far as really big, AOL is using Rails for the frontend of Revolution Health, rewriting the UIs of various acquired apps in Ruby.. im sure theres otheres, these are just the ones im aware of from reading this list, and job interviews etc..
> However if you are looking for names that everyone knows. Like ibm.com > or microsoft.com. Then no. There are no names like that.People havent heard of AOL, Amazon? even IBM and Apple have some articles up on their sites. in Apple''s case this probably grew out of the desire to sell more hardware to alpha geeks, but in IBM''s case it was likely extracted from usage scenarios.
On Tue Jul 18, 2006 at 01:33:16AM +0000, Jon Gretar Borgthorsson wrote:> A good list is here.. > http://wiki.rubyonrails.com/rails/pages/RealWorldUsage > > Among the big names are odeo.com, penny-arcade.com, calendarhub.com, > writertopia.com, streeteasy.com and so on. Most of these have traffic > that would make the average sysadmin blush.i doubt that. theres praobly more people that have blogged about these sites than there are dedicated users. care to Alexify your statement?
Jon Gretar Borgthorsson
2006-Jul-18 01:49 UTC
[Rails] Re: Rails - where are the BIG web apps?
On 7/18/06, carmen <_@whats-your.name> wrote:> On Tue Jul 18, 2006 at 01:33:16AM +0000, Jon Gretar Borgthorsson wrote: > > A good list is here.. > > http://wiki.rubyonrails.com/rails/pages/RealWorldUsage > > > > Among the big names are odeo.com, penny-arcade.com, calendarhub.com, > > writertopia.com, streeteasy.com and so on. Most of these have traffic > > that would make the average sysadmin blush. > > i doubt that. theres praobly more people that have blogged about these sites than there are dedicated users.I think penny-arcade has a few readers.... -- -------------- Jon Gretar Borgthorsson http://www.jongretar.net/
carmen wrote: On Tue Jul 18, 2006 at 01:33:16AM +0000, Jon Gretar Borgthorsson wrote: A good list is here.. http://wiki.rubyonrails.com/rails/pages/RealWorldUsage Among the big names are odeo.com, penny-arcade.com, calendarhub.com, writertopia.com, streeteasy.com and so on. Most of these have traffic that would make the average sysadmin blush. i doubt that. theres praobly more people that have blogged about these sites than there are dedicated users. care to Alexify your statement? Well, that wasn''t my claim, but... http://www.alexaholic.com/odeo.com+penny-arcade.com+calendarhub.com+writertopia.com+streeteasy.com Penny-arcade is the obvious winner here, with reach sometimes approaching 1K/1M and top-1000 rank. That would make all but about 1000 sysadmin blush, I guess. Odeo is a distant 2nd and the others aren''t even close (although streeteasy in particular is aimed at NYC only, so it''s hard to imagine they care about alexa reach). phil -------------- next part -------------- An HTML attachment was scrubbed... URL: http://wrath.rubyonrails.org/pipermail/rails/attachments/20060718/13a51ab0/attachment.html
Ken McMyre wrote:> Ok, Rails has been out for some time now....> Please, I''m not trying to flame - its just that these are the same web > apps I showed my boss 6 months ago and I certainly don''t want to go back > and show the same ones. Surely there are more large scale RoR web apps > out there by now? I see PHP powering a ton of big sites, I was kinda > expecting RoRs to have a little more by now.I think the large scale apps are under development, soon to roll out, or just too small to be well known yet. By the time they are large, the competative advantage to using RoR will have eroded because anyone in a highly competative space will be using it or something as close to it as possible in the Java/PHP world. Big companies and well-known brands are conservative when it comes to technology choice. They will wait for the technology to mature while their techies use it for internal and side projects. Startups are in a better position to take advantage of RoR and naturally they are unknown. I have more or less bet my company on it and I''m surely not the only one. Talk to me in six months. Steven -- Posted via http://www.ruby-forum.com/.
On 18 Jul 2006, at 22:00, Steven wrote:> Ken McMyre wrote: >> Ok, Rails has been out for some time now. > ... >> Please, I''m not trying to flame - its just that these are the same >> web >> apps I showed my boss 6 months ago and I certainly don''t want to >> go back >> and show the same ones. Surely there are more large scale RoR web >> apps >> out there by now? I see PHP powering a ton of big sites, I was kinda >> expecting RoRs to have a little more by now. > > I think the large scale apps are under development, soon to roll > out, or > just too small to be well known yet. > > By the time they are large, the competative advantage to using RoR > will > have eroded because anyone in a highly competative space will be using > it or something as close to it as possible in the Java/PHP world. > > Big companies and well-known brands are conservative when it comes to > technology choice. They will wait for the technology to mature while > their techies use it for internal and side projects. Startups are > in a > better position to take advantage of RoR and naturally they are > unknown. > > I have more or less bet my company on it and I''m surely not the only > one. > > Talk to me in six months.Indeed, also keep in mind some rails applications out there are used in private environments (i.e. the general public doesn''t see them). We will be starting development on a very large application (code- wise) in the coming weeks, but nobody will ever hear about it because it''s customized development and will be used in a closed environment. Already we tested performance of rails for a few of the critical components (with lots of hits, selects, creates, updates, deletes on very large datasets) and rails stood up very well and it wasn''t even optimized with caching. And all of that on a simple PowerBook G4 867, just imagine what it will do on a dual Xeon dedicated server. I have faith in rails as do my co-workers and things keep getting better, so... Best regards Peter De Berdt -------------- next part -------------- An HTML attachment was scrubbed... URL: http://wrath.rubyonrails.org/pipermail/rails/attachments/20060719/acf7125c/attachment.html
When considering scalability it is worth considering Gall''s law
A complex system that works is invariably found to have evolved from a simple
system that worked. A complex system designed from scratch never works and
cannot be patched up to make it work. You have to start over with a working
simple system.
Rails keeps things simple, so adheres to this law.
----- Original Message -----
From: Peter De Berdt
To: rails@lists.rubyonrails.org
Sent: Wednesday, July 19, 2006 11:50 AM
Subject: Re: [Rails] Re: Rails - where are the BIG web apps?
On 18 Jul 2006, at 22:00, Steven wrote:
Ken McMyre wrote:
Ok, Rails has been out for some time now.
...
Please, I''m not trying to flame - its just that these are the
same web
apps I showed my boss 6 months ago and I certainly don''t want to
go back
and show the same ones. Surely there are more large scale RoR web apps
out there by now? I see PHP powering a ton of big sites, I was kinda
expecting RoRs to have a little more by now.
I think the large scale apps are under development, soon to roll out, or
just too small to be well known yet.
By the time they are large, the competative advantage to using RoR will
have eroded because anyone in a highly competative space will be using
it or something as close to it as possible in the Java/PHP world.
Big companies and well-known brands are conservative when it comes to
technology choice. They will wait for the technology to mature while
their techies use it for internal and side projects. Startups are in a
better position to take advantage of RoR and naturally they are unknown.
I have more or less bet my company on it and I''m surely not the
only
one.
Talk to me in six months.
Indeed, also keep in mind some rails applications out there are used in
private environments (i.e. the general public doesn''t see them).
We will be starting development on a very large application (code-wise) in the
coming weeks, but nobody will ever hear about it because it''s
customized development and will be used in a closed environment. Already we
tested performance of rails for a few of the critical components (with lots of
hits, selects, creates, updates, deletes on very large datasets) and rails stood
up very well and it wasn''t even optimized with caching. And all of that
on a simple PowerBook G4 867, just imagine what it will do on a dual Xeon
dedicated server.
I have faith in rails as do my co-workers and things keep getting better,
so...
Best regards
Peter De Berdt
------------------------------------------------------------------------------
_______________________________________________
Rails mailing list
Rails@lists.rubyonrails.org
http://lists.rubyonrails.org/mailman/listinfo/rails
-------------- next part --------------
An HTML attachment was scrubbed...
URL:
http://wrath.rubyonrails.org/pipermail/rails/attachments/20060719/0dd3ff26/attachment.html
Ken McMyre wrote:>Big companies and well-known brands are conservative when it comes to >technology choice. They will wait for the technology to mature while >their techies use it for internal and side projects. Startups are in a >better position to take advantage of RoR and naturally they are unknown.Hi Ken, I tend to disagree with this statement somewhat. In our experience running workshops we''ve seen many large companies attend our Rails training specifically. For instance the last Rails workshop we did in London was attended by the likes of American Express, BBC, Rolls Royce and Reuters. And they weren''t sending just one person but in some instances four or five people. I think that''s very encouraging for Rails. I know that these companies don''t have any web apps to show for this but they are definitely embracing the technology, or at least plan to very soon. Just thought you''d be interested. Gill gill@carsonworkshops.com -- Posted via http://www.ruby-forum.com/.
Gillian Carson wrote:> Ken McMyre wrote: > >>Big companies and well-known brands are conservative when it comes to >>technology choice. They will wait for the technology to mature while >>their techies use it for internal and side projects. Startups are in a >>better position to take advantage of RoR and naturally they are unknown. > > Hi Ken, > > I tend to disagree with this statement somewhat.That was me.. not Ken.> In our experience > running workshops we''ve seen many large companies attend our Rails > training specifically. For instance the last Rails workshop we did in > London was attended by the likes of American Express, BBC, Rolls Royce > and Reuters. And they weren''t sending just one person but in some > instances four or five people. I think that''s very encouraging for > Rails.I am sure that there are many internal projects. And companies like you mentioned have budget to send their techies to training for whatever-is-hot whether or not they are actually using it.> I know that these companies don''t have any web apps to show for this but > they are definitely embracing the technology, or at least plan to very > soon.I''m sure they are!> Just thought you''d be interested.I think there is some confusion regarding what is meant by a "large-scale" app. The original poster seemed to be looking for something he could point his management to in order to increase their comfort factor that someone out there is using RoR successfully in a "large-scale" app. [To anyone:] What would qalifiy as large-scale in your mind? My current site is used within a professional community and has 10k+ registered users. It is not a consumer site yet and I would consider this small. But it has to scale to 10x that or more within the next year. I am comfortable converting this to RoR. A couple years ago, I consulted for an also-ran in online travel and they were doing about 1 million searches and 3000 bookings per day on a messy Java/ Perl/ Oracle application. It was everything we could do to keep it stable and running on expensive and beefy hardware yet those were safe and "mature" technologies at the time. I''ll bet the whole thing would have worked much better on RoR if it had been around. It would have required 3-4 programmers instead of 10+ and much less hardware. Of course their problem was that most of the programmers learned on the job, building that as their first major system. That is not a recipe for stability and scalability no matter what platform you are on. Everyone has to learn somewhere but I believe experience does count for something. Steven -- Posted via http://www.ruby-forum.com/.
Steven wrote:> keep it stable and running on expensive and beefy hardware yet those > were safe and "mature" technologies at the time. I''ll bet the whole > thing would have worked much better on RoR if it had been around. ItHaving been in I.T. professionally for more than a decade, I''ve long noticed the corporate bias in favor of "mature" technologies, as though there''s something magical about the fact that tools are older. Of course, the newer, more "immature" tools are developed precisely because of perceived shortcomings in the older tools, or because alternate ways of looking at problems have created new solutions. To the extent that newer tools are developed to overcome perceived shortcomings in the older ones, that seems to me a good thing, not something to be fearful of. If "mature" technologies were always the panacea that some think they are, I suspect we''d all be writing Web apps in COBOL or assembler. -- Tammy
Tammy Cravit wrote:> Steven wrote: > >> keep it stable and running on expensive and beefy hardware yet those >> were safe and "mature" technologies at the time. I''ll bet the whole >> thing would have worked much better on RoR if it had been around. It > > Having been in I.T. professionally for more than a decade, I''ve long > noticed > the corporate bias in favor of "mature" technologies, as though there''s > something magical about the fact that tools are older. Of course, the > newer, > more "immature" tools are developed precisely because of perceived > shortcomings in the older tools, or because alternate ways of looking at > problems have created new solutions. To the extent that newer tools are > developed to overcome perceived shortcomings in the older ones, that > seems > to me a good thing, not something to be fearful of. > > If "mature" technologies were always the panacea that some think they > are, I > suspect we''d all be writing Web apps in COBOL or assembler. > > -- TammyThe "something magical" is that a decision maker has to cover his/her butt when making a decision. Executives get promoted for good choices and fired for bad choices. To minimize risk, if they go with the gorilla/market leader then they reduce some risk in making technology decisions. It is the PERCEIVED safe path, not necessarily the best path. Small, unproven technologies come and go and very few high-up decision makers want to hedge their bets on the survival of something that they are investing in. I''m not suggesting that Rails is "small and unproven" - as I absolutely love the whole framework and developed an internal CRM application for our company. I am suggesting that the person who started this thread is seeking validation from someone, somewhere so that he can SELL this to his management. He already loves the framework - now he is trying to help sell it and using the most powerful tool available which is a proven large account. Just my opinion! I sure am eager for this to continue adoption in the corporate world as I could use the consulting $$$! :-) Michael mmodica at cox dot net -- Posted via http://www.ruby-forum.com/.
> The original poster seemed to be looking for something he could point > his management to in order to increase their comfort factor that someone > out there is using RoR successfully in a "large-scale" app. > > [To anyone:] What would qalifiy as large-scale in your mind? My > current site is used within a professional community and has 10k+ > registered users. It is not a consumer site yet and I would consider > this small. But it has to scale to 10x that or more within the next > year. I am comfortable converting this to RoR.I would love to be able to tell you the site, but I can''t just yet. In a month I''ll be able too, I promise :) Right now I''m responsible (well, not *just* me) for a site spread across 11 front end servers. And 90% of it is in Rails. A couple (small) pieces are still PHP, but only cause we haven''t converted them over. About a week ago we did 3,697,616 pages, 28,294,940 hits, and 255.72GB of bandwidth. For July we''ve avereged just under two million pages a day. There is only one cached page (the home page) since the content changes too quickly to make page caching useful. I ran some numbers on the busiest time during that day and figured we can easily handle 61,954,728 pages per day with our setup -- not that we''ll need to, but we could. -philip