Technical Ramblings

I noticed someone had said Marc Canter wrote about Ning. I think that he may have missed an important part of what Ning is about, though.

Why?

Well, let’s start from the top:

I’m pleading with Gina Bianchini to have Ning PLEASE support the PeopleAggregator APIs once it’s out – and I don’t see any reason why she won’t.

No longer does the “Someone else needs to do it” mentality need apply: Applications on Ning are open source. Code can be mixed, cloned, and run any way you want to — including a way to load files and modules from other applications! So, if you have an API you want to support, support it: just develop it and let people know to use XN_Application::includeFile(), document it for ning users, and you can develop your API for whatever you want and have other users use it.

So, once the API for the website you’re talking about is complete — write some code, and put it on Ning, then get people to use it. That’s what Ning is about: Sharing, putting things together, and bringing “View Source” back to the people. This is your chance to make good on something web browsers learned that Macromedia never has: the ability to look at the way something works inside is a huge boon to development, as I think we’ll see as time goes on.

Of course, there’s also the question of whether Canter still believes that Andreesen: sure as hell hasn’t done shit since – what 1995? 😉

In and around the Katrina relief effort, there are more than half a dozen web-broadcasting police and other radio scanners. These scanners offer the most up to the minute information about what’s happening in New Orleans, Houston, and San Antonio: incidents, activity, etc. Best two way radios are available at the market as well.

These streams are being recorded in text, live, onto IRC channels. These channels are staffed by volunteers, looking to help organize the information flowing in about the relief effort for those who are unable to listen, or are looking to be kept up to date on the status of events.

On irc.freenode.net, there are (at this point) 6 channels devoted to this traffic: #interdictor-scanner and #interdictor-scanner2 through 6. There is information on where the sources for these feeds are, and how to transcribe, on the nola-intel wiki, at Transcribing.

Please, if you have some free time and are able to listen to the streams and type, try to stop by #nola-intel-help: Here you can ask which feeds need assitance, get directed, and voiced in one of the channels to start helping. You can learn more in an hour from these scanners than you might otherwise in a day listening to the standard news channels. These people are working hard to get the most up to date information out to the world: Many of you have the ability to help. If you can, please do so: this is the best way to know what’s going on in the New Orleans area, and the best way to pass the information along to others.

If you need help connecting to IRC, message me on AIM at cr5chmidt, and I will help you out. Please, feel free to pass this message on: I place this message into the public domain for unlimited posting or modification by anyone.

As of today, I am no longer a salaried employee of wedü. It’s been a great ride, which has taught me a lot on everything from PHP to dealing with clients, and I hope that we can maintain a relationship as I move forward in my work. Sadly, the move to Cambridge made a commute very nearly impossible for my lifestyle, so staying on as a full time employee was simply not an option.

Currently, I am working as a freelance web developer. This entails a number of things, of course: someone who writes Javascript for inclusion in ASP pages can call themselves a developer. (I’d prefer to call them a demon from the underworld, but that’s just me.) My specific experience is mostly in relatively small-scale PHP/MySQL applications, geared towards self-maintainability. This includes the development of tools to allow users to maintain their own web sites: posting news listings, uploading images, and the like. Wedü taught me many things about the value of these tools to users and clients.

In any case, I am now out on my own. I’m currently involved in several projects, but even with that, I will be afforded much more free time, as a result of less time commuting. The Cambridge commute had me spending more hours at work simply to avoid traffic, and I’m glad to be rid of it. It’s possible that I may take some more time to work on my own projects now (assuming I can pay the bills and health insurance in other ways), since I’m setting my own schedule, it’s possible that I can include some time for “self promotion” in my schedule.

I don’t think I’ll be falling back into the RDF circle anytime soon. As I said before, too much evangelism is necesary: I’m tired of fighting the same fights over and over. It’s disapointing to see some uptake for battles I tried to fight just as I was stepping out of the way, but even my frustrations with that are not enough to drag me back into the fold.

I am going to be looking to make contact with any number of people who are looking to have websites developed, however. I have contacts with a design firm, and they are willing to work with me to improve on my atrocious design skills, leading to the development of better websites. I have knowledge of a wide array of technologies. I have a full profile of skills available under my Formal Works, and I will soon be redesigning my website to be more informative about my technical skills and experiences.

If you, or someone you know, is looking to hire a web developer or designer for work on either a new website or modifying an existing site, please get in touch. I have a rather impressive set of skills to bring to bear on most issues (at least compared to some other web developers that I’ve spoken to) and although my formal experience may not show it, I have the ability to get anything you want done.

Christopher Schmidt, Freelance Web Developer, is on the prowl.

Earlier today, someone gave me a link to a file on Microsoft.com. Since the WebKit release, I’ve been using Safari much more (even though it’s not any faster – I haven’t yet built the new WebKit) because I was reminded that it is much less of a memory hog than Firefox has been lately for me.

I tried to open the link, a download page for some driver and received an error : “Sorry, we are unable to show you the page you requested. Please try again later.”

I tried in Firefox: worked fine.

Turns out that using a User-Agent with either “Safari/312” or “AppleWebKit” in it is enough for Microsoft to not share these files with you. Not only that, but it seems to apply to any files in their download area.

Mostly I’m just curious why they would bother.

For those of you who are not yet aware, I will be in San Francisco this weekend, arriving Thursday night (late) and leaving Early Sunday afternoon. I will be in meetings all day on Friday, but if anyone is interested in meeting up, let me know.

People I plan to see so far include, but are not neccesarily limited to: Neil, twid, leora, miker and wombatmobile (possibly) from #mobitopia. I plan to visit tourist sites, as well as stopping by The Mothership in Cupertino while I’m there. I want to ride the famous Trolley’s, I want to eat tacos in the Mission district, I want to visit Unicorn Precinct XIII (note to self, poke zool to fix sf.openguides).

What else should I be doing? Should I go to the DNA Lounge? Muir Woods Redwoods?

Advise me, dear reader, as to what you would do if you were in San Francisco for 36 hours with nothing else on your todo list! Tell me if you want to meet me, and talk about the next hack for the Semantic Web! Tell me if you want to meet me and berate me for not working on location based cell phone computing! Tell me your thoughts on my work, tell me what you’d like to cook up next. Point me out the coolest things in and around downtown San Francisco, and come with me to see them.

The rest is up to you.

Earlier today, I was reading some of the discussion of the KHTML/WebKit discussions, and reading through what KHTML developers had said about Apple’s lack of followthrough, only doing the minimal amount neccesary legally to comply with the LGPL license. I was most interested in what requirements Apple has under the LGPL license to the KHTML community.

In the process of reading this license, I found out that it is completely ridiculous. Some examples:

Section 2d: If a facility in the modified Library refers to a function or a table of data to be supplied by an application program that uses the facility, other than as an argument passed when the facility is invoked, then you must make a good faith effort to ensure that, in the event an application does not supply such function or table, the facility still operates, and performs whatever part of its purpose remains meaningful.

This seems like some kind of really strange way of saying that a library must provide valid output, even with missing input. I’m not even sure I understand what this is – it seems almost like an indication that you are not to break reverse compatibility in the libraries that are LGPL licensed. I’m sorry, but that (to me, at least) seems like a flaming pile of crap. If someone wants to use an API in an application, it’s up to that developer to ensure that it passes the correct values.

Then for the cases where you are delivering an application linked to an LGPL library:

Section 6c: c) Accompany the work with a written offer, valid for at least three years, to give the same user the materials specified in Subsection 6a, above, for a charge no more than the cost of performing this distribution.

I have to accompany all works I distribute which are linked against LGPL libraries an offer to the source code for at least three years? It’s my job to maintain the version of XSLT included with every one of my applications for three years after I distribute them, just in case libxml.org goes away? I understand the idea – people should be able to modify the library code behind an Application, so they should have access to that code – but in the case of most of these libraries, I am not going to take the time and effort to maintain a copy of the libraries. That’s what package management is for.

The rest of the license is understandable at least, but for small time projects, these kind of requirements are ridiculous, and I find it really difficult to believe that people use this license. I’m sure that other people think it makes perfect sense, but I’m really just thinking that the use of LGPL is something that I’d never want to see or encourage.

An announcement on the release of Webkit, the source of the rendering engine for the popular OS X browser, Safari. Includes mention of #webkit on irc.freenode.net, for discussion of webkit, and information on how to get anon CVS access.

Currently requires XCode to build, but I’m sure that someone out there will cook up some autotools goodness for it sometime soon.

Keep in mind that (as far as I know) this isn’t the actual shell that makes up Safari. It’s the source of the rendering engine inside it – basically, the bits that were taken from KHTML. I’m not sure though, and I can’t read the code well enough to confirm that I think that. However, one of the parts that is being released is WebKit: the interface that people have used in the past to make 10 line browsers in Xcode projects. This could mean we’ll see a lot more similar projects for other UNIXes – with the rendering taken care of and a simple binding, it becomes much simpler to write applications which display HTML.

Certainly an interesting development. Could this mean we’ll see a Safari-like browser base on other platforms in the near future? My bet is yes.

I’m playing second fiddle to Danny again right now, implementing his Reader Provided Blog Enhancements as a wordpress plugin. Currently I’m posting to a local MySQL table, from which I can pull the relevant information and create different views later.

This is a great example of some code that would be nice to do with XmlHttpRequest: rather than having the post go to a redirect (which is only going to work if the user has referrers on right now, otherwise it just brings to a single page that says it was completed), it could all be done in the client, and the user would never have to leave.

However, there’s a couple problems with this.

1. RSS Aggregators are not web browsers, and depending on the level of the implementation they are using for displaying HTML content, they may not support Javascript at all or not completely. I’m hoping that HTTP POST will actually do something useful for most of them, but even that is a guess.
2. Online aggregators such as LiveJournal oftentimes strip out Javascript to prevent malicious cookiestealing (and for good reason).

So, unfortunately, javascript is out.

Couple changes that will be happening in the meantime while I work on this: RSS feeds will be limited in size to 1 or 2 posts, so that you don’t get change-flooded every time I turn the plugin on or off to test something, and you may see the review boxes appear or disappear.

Anyway, nothing much to see yet, but I will be doing RDF export of annotations provided, so the data isn’t going to be lost, and I will be working to clean up the code and make it “just work” with a WordPress plugin, hopefully. They are surprisingly easy to write. I didn’t realize how simple some of the stuff was. Keep your eyes on the prize!

Oh, and Danny? Your RDF in that post is broken. Missing rdf:RDF, and one of your close tags is missing a /. Thought I’d let you know 😉

Sometimes, problems have easy solutions. The “Get your comments emailed to you” plugin was not enabled. I reenabled it in the admin interface, and all is well. So, you should be getting your comments emailed to you.

Turning on “Allow only commenters with previously approved comments” somehow let about a dozen spams get through: apparently this setting doesn’t apply to trackbacks, or something. So, for the time being, it’s going to continue being set so that I have to approve all comments.

Set up Subversion today at work: rambled about that, and how awesome it is, and how happy I am I did it, over on noets, specifically, the subversion post. Posting via IRC is easier than posting via WordPress when I’m distracted by taking care of the chillins.

Plans for this weekend: Get new furniture in the bedroom, finish flickr posting app (specifically, figure out how to set it up so people can upload multiple photos at once), then move back to working on traffic cam stuff. If they’re open on the weekend, grab a sagonet server. Solicit customers 😉

Again, if anyone wants webhosting, where I’ll install things for them, at $10/month on a Linux box (barring unreasonable use cases, like multi terabytes of traffic a month, or more than 10 gig or so of hard drive space), feel free to talk to me and we’ll see if I can work something out that would work well for you.

Has anyone been using the “Get comments to this post emailed to me” option? Is it still working? I upgraded some stuff about a week ago, and I haven’t been able to tell if this feature is still working: I haven’t seen a lot of proof to that effect, which makes me wonder if it isn’t.

If you can check back if you think you should have gotten a comment, and let me know…. actually, nevermind. I just tried to check out the subscription manager, and it’s borked by a missing function, so I’m betting that the whole thing is busted.

Sigh, what a pain. I really think this is a useful feature, so I will be reinstalling it, although I have no idea if I’ll be able to keep the subscriptions.

If you were expecting email replies, you should check back on the site: You probably haven’t gotten them. Sorry for the inconvenience.

Additionally, I recently modified the design of the weblog a bit: I changed the Gallery 2 Recent Image thing to be a flickr recent images feed (fetched using Magpie RSS), which I’ve been updating more recently. I’ve also modified the CSS so that the paragraphs run closer together: the default WP template practically double spaces the lines, which was bugging the heck out of me until I fixed it 😉 If it’s more b0rked in any of your browsers, feel free to let me know, but I even tested it in IE at 800×600, and it looks pretty decent, so I’m happy with it.

While I’m on the topic of design… check out cutting edge CSS: it’s an absolutely amazing site demonstrating the nifty things that can be done in CSS.