Posted by Kevin
Sun, 25 Jun 2006 19:05:00 GMT
- The nubyonrails.com guy
- This session is about therapy. If you have had problems with shared hosting in the past, it’s not your fault.
So why do people choose to use shared hosting? It’s cheap and it’s easy.
And why is it such a big struggle? There are many pigs sucking at the same teet.
Recommendations
Troubleshooting
- The same stuff always goes wrong
- Check you shebang line (the top line of any ruby script, especially in dispatch.fcgi)
- ./script/server to see if the app can even load in webrick
- ./script/console to see if you can even hit your database
- tail -f your logs
- And if none of that works…just wait, someone else may be screwing with something
Buy my book: Rails Deployment from the pragmatic programmers.
This Geoffrey, representing properly.
Posted in Railsconf | no comments | no trackbacks
Posted by Kevin
Sun, 25 Jun 2006 18:30:00 GMT
- Red Sox Fan, PhD from MIT, works for Verisign, country music fan, from India, not an alcoholic
- In the real world, people see me in different ways depending on the context: family, friends, business, hobby.
- On the web, I’m put into a box. Each site requires me to enter my personal info in the way they want, and I must remember each of my usernames and passwords.
- This is site-centric, not user-centric
- When traveling, an airline asks for and accepts any state driver’s license as identification. The driver’s license is interoperable, multi-purpose and portable.
- GET ME ONE OF THOSE FOR THE WWW!
An Open Identity system should be…
- Extensible
- User managed
- Open
- Perpetual (not just used once)
Sites already on board:
- Livejournal, wikimedia, zoomr, schtuff, railsweenie
It’s about you
- The general idea is that when you go to a site, you are redirected back to your site, where you log in. The site that you intended to go to communicates to your site and confirms your identity, allowing you into the site you wanted to go to in the first place. And best of all, once you do login once, you’re on: Single Sign On!
- There are attributes that you control. You can specify the attributes that you want to share with each site.
- Some attributes (such as age, like if you are on a site selling alcoholic beverages) need to be verified. Certain companies would then be in the business of verifying OpenID attributes.
Other Technical Presenter Guy
- The CTO of verisign went through a setup of an OpenID client in rails. Pretty hot.
- Some sites may require certain attributes, so there may be opt-in/opt-out. A lot of the parts of the implementation are still under discussion.
- There is a serious question about how sites will know which OpenID servers to trust. It is neat to think that you could have your information hosted on your own website, but who are you anyway, that you would be trusted? There has been talk of servers being trusted if they are a “friend of a friend.”
Awesome stuff, if it ever works out in practice. Jeffmo is skeptical. Matt Pelletier from eastmedia has been working on this as well, and he gave a talk on the topic Sunday morning.
Posted in Railsconf | no comments | no trackbacks
Posted by Kevin
Sun, 25 Jun 2006 17:10:00 GMT
“Hello, I’m David. I’m here to tell you what you’re doing wrong.” (Yeah, good solid shock value there)
He has looked at websites since ‘95. Used to work in tech support, so he’s heard lots of customers talk about what’s not so good about various interfaces.
What goes wrong?
- While we (web 2.0 developers) are all excited about the “new” web, much of our potential audience is still discovering the old web.
- We assume that our users control their environment: if they have and windows machine with internet explorer, that’s because they chose it.
- In the real world, people are not aware of the “new hotness”, they don’t think they need it, and it has not been properly explained.
- User requests (for features, etc) are often “aspirational”; which is to say that they are something the user would like (a million dollars) rather than something they need
What can be done?
- Find out about your customers; talk to them.
- Don’t work towards a comprehensive feature list. Too many features can do more harm than good.
- Think about Usability through your site’s audience: Who’s going to use it? How are they going to use it? How often will they use it?
Syndication and Tagging
- 12% of internet users are aware RSS exists
- 4% have knowingly used it
- Folks usually just have tiny RSS icons and not much explanation, but these kinds of stats make it seem reasonable to expect web authors to explain how to use RSS (Or hot new technology X) on their sites.
- We know and love tags, but we can’t assume that people will know how they are used (or how they are used in our app).
- Think different – he called out one site that uses tags to review wines.
Fair enough, rock on.
Posted in Railsconf | no comments | no trackbacks
Posted by Kevin
Sat, 24 Jun 2006 22:01:00 GMT
The presentation started off with renditions of “You Shook Me All Night Long” and “Enter Sandman” on the accordion, which was, obviously, the hottest thing ever
“How the hell am I going to pull off a comparison of AC/DC Stravinsky, and Rails?”
AC/DC
- Adams Law of AC/DC: You know the title, you know the chorus. This works well with the rails concept of rails routes, whereby knowing the URL tells you a lot about where to the find the code for that page.
- AC/DC doesn’t add unnecessary notes (no really, there are very few notes), and rails doesn’t make you add unnecessary code.
- AC/DC is consistent. It’s strait up rock, and it’s always about sex or drugs or rock and roll. All rails primary keys are called id. ‘nuff said.
- Bus Factor (when AC/DC lost their lead singer and got a new frontman, they proceeded to release one of their best albums – Back in Black). Rails also makes it easy to spread the knowledge around a programming team (by limiting complexity).
Igor Stravinsky – The Rite of Spring
- Takes 110 peope to perform (and we know how big a pain in the ass it can be to coordinate communication with two people).
- It isn’t necessary or desirable to emulate the complexity of The Rite of Spring when building your rails app.
- Part of the piece is in 11/4 time, which is a bit nonconformist, as rails has historically been as well.
- A New York Taxi driver joke goes, “How do you get to Carnegie Hall?” (Answer: practice) It takes about 1000 years of total training time to get people equipped to perform The Rite of Spring. Yeah, again, don’t build software like that.
All of these things (AC/DC, The Rite of Spring, and Rails) have a tendency to cause problems. The dancing in The Rite of Spring caused chairs to be thrown, and rails leads to flame wars. This is because all of these things incite passion. While Microsoft may say, “we need people who are non-threateningly dressed to build great solutions,” the rails community understands that value comes from the margins. AC/DC has created, Stravinsky created, and rails allows us to create. We’d like to think that, like AC/DC and Stravinsky, our work is not a discipline, but a craft.
Big Balls was then played on the accordion. I take it back, that was the hottest thing ever
Posted in Railsconf | no comments | no trackbacks
Posted by Kevin
Sat, 24 Jun 2006 21:28:00 GMT
Dave Thomas is the hot pragmatic programming guy apparently. His points follow:
Good news everybody!
- People are using rails to actually do business!
- The trendline for rails is going in the right direction
- Matz in the man
- Core is the man (or something)
But…There are still things to do. Unsolved problems according to Dave:
Data Integration
- It would be neat if validations were added to models automatically from the schema
- Foreign key support should be better (good point)
- Primary keys are always integers? Is that cool?
- Can we have composite primary keys please
- Can we get better support for non-database-backed models
Real World Crud
Scaffolding brings people to rails. Why not improve it a bit? Rails is supposed to be the poster child of rails 2.0, but the current scaffolding is the worst of web 1.0.
- Make it configurable (allow radio button, check boxes, etc)
- Show off – use Ajax
- Allow skinning.
Deployment
“Who has deployed a rails app,” he asks. (everyone raises their hand) “Who enjoyed it?” (like three people raise their hands)
So suggesting these kinds of things for inclusion in rails core raises the question, why? Well, people use rails because it makes their lives easier. The rails community places a high premium on writing code that others can use. If we can help to lower the bar in terms of barrier to entry for rails, we can make other peoples’ lives easier. “All developers deserve to be happy.”
Good talk man – solid beginning for Railsconf.
Posted in Railsconf | no comments | no trackbacks
Posted by Kevin
Sat, 24 Jun 2006 21:14:00 GMT
Right, so I’m at the First International Rails Conference in Chicago from June 23-25. I’ll probably be dropping a good number of notes about the same in the next couple of days, mostly in case anyone at work asks about what I was doing at 1:30pm on Friday (I don’t plan to remember otherwise).
Posted in Railsconf | no comments | no trackbacks
Posted by Kevin
Sun, 21 May 2006 23:33:00 GMT
I believe that releasing new software when it is done1, and immediately when it is done, is a tremendous boon to the developer (and by definition the customer, unless you happen to be adding features the customer doesn’t want). I’ve dreamed about this a lot, and advocated for daily web rollouts at work for months, so naturally I was pleased to find that Paul Graham has argued this point in his essay, The Other Road Ahead.
There seems to be much opposition to this idea in the software arena. Releases are often no fun, and many seem to think that releases are inherently difficult, time-intensive and error-prone – and should therefore be minimized. Some say that customers need to be conditioned not to expect what they want now. There is definitely truth in this. But for all of the reasons releases are tricky, I believe they should be done more and more incrementally.
Releases are difficult
Sure. Releases will be difficult as long as:
- People assume that they happen only infrequently
- There is a perception that it just isn’t worth the effort to make them work the way they should
If I used the vim text editor once a week, I wouldn’t bother creating/learning the cutesy shortcuts. Institute a continuous release policy and the process will improve guaranteed.
Releases are time-intensive
Yes, if there are many new features, QA can become a long and uncertain process. And when a bug is found, the amount of time that has passed since the code was written correlates fairly directly to the time that will be needed to fix the bug.
Releases are error-prone
Certainly, managing the complexities and anticipating the interactions between features of a large release is an unenviable task.
So if new code is ready, or if a feature the customer wants is finished to everyone’s satisfaction, I feel that there shouldn’t be a ton of hoops to jump through before the new software finds its way into production. Developers will ultimately be happier to see their creativity in finished form faster, customers will get their feature more rapidly, and potentially, delicious manna will pour down from heaven.
1 Done, in this case, means ready for release, including any necessary testing.
Posted in Software, Philosophy | 2 comments | no trackbacks
Posted by Kevin
Sat, 06 May 2006 02:31:00 GMT
I read somewhere that “The Fair and Accurate Credit Transactions Act (The FACT Act) entitles all Americans over the age of 18 to receive one free credit report from all three credit reporting agencies.” I checked it out, so here’s wagwan...
You apparently have a right to know what’s going on with your credit, but you apparently don’t have a right to get your actual numerical credit score for free. Damn.
I believe one of the companies offered to give me a number for six bucks, but that seems pretty steep after I thought it was coming in the mail for free. On the up side, the reports are fairly complete, completely free (once a year), reasonable straitforward – and they came about three days after I called in the request.
You can check out the service online or use their toll free number, 1-877-FACTACT, to hook yourself up.
Posted in Finance | Tags Credit, Finance, Report | 2 comments | no trackbacks
Posted by Kevin
Sat, 22 Apr 2006 07:18:00 GMT
It’s the third day this blog has been in production, and as such, it’s about time to wax the default Typo theme.
Azure, as it’s dubbed, was pretty hot a year ago , but these days there’s so many kids using it that it just doesn’t seem necessary to add one more. The new theme is still mostly stolen from the same, but I went ahead and blasted the color scheme and made the layout fluid in the process.
Posted in Typo | no comments | no trackbacks
Posted by Kevin
Thu, 20 Apr 2006 06:13:00 GMT
Well, it kind of seems as if Typo is working with the rails of the day. Sweet.
Posted in Rails | 1 comment
