People jumped on javablogs for their recent outage.
The javablogs developer at Atlassian has written about what is going on at javablogs.
The item I appreciate is:
Language Filtering: English Only | All Languages
Javablogs, due to popularity, was getting blogs in many many languages. This is great. However, since I can’t understand most of them, it was a royal PITA to wade through.
It would be nice to see a tiny open source component that does the
content.languageIs(ENGLISH) piece so it can be shared, and tweaked to be better and better.
I want this on technorati search results too.
The No Fluffers have been discussing the Lifehacker article on avoiding broken links.
Is this just a hack that email clients agree to grok? Glenn Vanderburg found that this isn’t the case, and that it is in the RFC for handling URLs (RFC 1738)
APPENDIX: Recommendations for URLs in Context
URIs, including URLs, are intended to be transmitted through
protocols which provide a context for their interpretation.
In some cases, it will be necessary to distinguish URLs from other
possible data structures in a syntactic structure. In this case, is
recommended that URLs be preceeded with a prefix consisting of the
characters “URL:”. For example, this prefix may be used to
distinguish URLs from other kinds of URIs.
In addition, there are many occasions when URLs are included in other
kinds of text; examples include electronic mail, USENET news
messages, or printed on paper. In such cases, it is convenient to
have a separate syntactic wrapper that delimits the URL and separates
it from the rest of the text, and in particular from punctuation
marks that might be mistaken for part of the URL. For this purpose,
is recommended that angle brackets (”<” and “>”), along with the
prefix “URL:”, be used to delimit the boundaries of the URL. This
wrapper does not form part of the URL and should not be used in
contexts in which delimiters are already specified.
In a little testing it appears to work in a fair few email clients. The only problem I have had was in Gmail for Mobile (Cedric will fix that I am sure ;)
Of course, you can always use tinyurl.com, xurl.us, and the myriad of other services.
We have finally gotten out west to the bay area. We are settling in, and work starts on Monday. I have been told to “bring your signing hand” for a fun day of HR ;)
We have overheard a couple of conversations from people moving themselves (on the plane, at Pete’s Coffee, etc):
- Us: The bad news is that the weather isn’t great here. The good news is that it has been good enough for me to be in a t-shirt, which is a touch nicer than 20 degree weather back in Wisconsin. We are looking to a sun upgrade.
- Rain Sucks: “Now I remember why I moved to Arizona. The weather here is terrible”
- Midwest: “We get all four seasons in the midwest, so I moved there from California”
Everyone has their own opinions. I am looking forward to exploring the red woods, pt reyes, wine country, head up to Oregan, down the coast to more sun, etc.
If you know of some hidden gems, let us know!
I signed up to Geni the new viral family tree site to play around on it for the Ajaxian post.
The viral piece really does work. I put in a few family members and when I went to my tree the next day I saw 400 entries in it! Family had added family, who were invited, signed up, and the circle continued. Suddenly I had my family tree and had not had to do much of any work :)
The Geni launch had been so successful that the site couldn’t keep up. The site was barely functional for a significant time, and even when it was up it was randomly slow.
If a Web 1.0 site is down, you know. The site doesn’t respond. No page comes up. People get that, go away, or hit Refresh continuously (often making the problem worse).
Geni has a rich Flash/Ajax interface and it wasn’t obvious to people when it was in trouble. The interface had been loaded and was fine but when you went to add someone it would wait for awhile. This made putting in content painful for family members but some thought it was an issue on their side.
The feedback is different to what they are used too. Click on a URL. No response. I get it. Cause an event in a rich UI that otherwise seems like it is working? not as easy to grok. “I can move my family tree around and zoom in, so it must be working. Why won’t it let me add a person though?”
I just noticed that Pactester has been released:
Pactester is a tool to test the proxy auto-configuration (PAC) files used by browsers to identify the correct proxy server for a given URL.
Since the PAC file evaluation mechanism is generated inside the browser and cannot be accessed from outside, the only way to tell which proxy your browser will use for a specific URL is manual inspection of the PAC file. Unfortunately, this approach is error prone and quickly becomes impractical for large and complex PAC files. Enter pactester, which resolves this issue by simulating browser behavior.
Normally I wouldn’t think much about this, but I am seeing PAC come up more and more, normally surrounding doing offline/sync work for Ajax applications.
I did a typical rant about the URL bar auto- suggest in Firefox, and how it can be annoying when you are in the use case of “most common” and it is throwing up “most recent”.
The beauty of being in our profession, and having a tool that is scriptable and extensible, is that you can do something about it.
Firefox also makes it simple, because you can get access to the URL bar via:
document.getElementById(”urlbar”); // aliased via gUrlBar in many contexts
I wrote the URL Bar Navigator plugin to do what I needed. A simple Control + Left Arrow narrows the scope of a URL that I am on to just the root domain, and I am enjoying it already.
Kudos to the Mozilla team for making life so easy. It isn’t without warts though. There are many resources on building Firefox extensions, but it is hard to grok which are out of date (FF 2.0 happy). The system is flexible, so you run into the ant-vs-maven problem of “hmm, should I create a chrome directory to put the overlay in? or should I just put it up here”. The extension extension helped a little, but it was buggy for me on Mac OS X. Many of the scripts didn’t work on OS X too.
Does this seem a little twisted?
I would think that poor JRuby would cry out and say “hey, just use me!”
At first it seemed obvious that I would get an iPhone. It is the phone I have wanted for years.
Yet, over the last couple of weeks as details come out about it, I am souring on it.
The biggest issue is that it may not allow any third party applications.
I expected the opposite Java, Objective-C, and even Ajax apps. A huge ecosystem would come about as you could now develop for the coolest phone (as well as others).
To hear Steve b.s. about third party apps not being worthy (killing the phone? hammering the network????) is painful and FUD.
I love my Apple products, but although I enjoy their software, it is third party software that really makes my Mac’s hum. Would I be on an Apple if I didn’t have Quicksilver, NetNewsWire, Textmate, insert the rest of this list?
No. There are always going to be better applications built outside of your company. Let the platform lose and see what amazing things people build Steve. If not, the iPhone will be eclipsed by copy cats that do.
The releases kicked in over in Rails land:
Haml you ask?
= stylesheet_link_tag 'main'
%p All content copyright © Bob
Watching all of the alternative view technologies gives me a bit of a Deja Vu to other technology. JHTML, JSP, Velocity, XMLC, Freemarker, ….
Before long we will have to fight against Enterprise Ruby Beans? ;)
Now that is a great bug. Getting your errors on packaging.
I can’t wait to see the Web services version of a 500 error on the front of the package.
(via Jim Halberg who I know “owe” for a nice send off last night)
And then Nick Sieger pointed to: