Author: matt

Original Ajax

I thought this was interesting over on Scobleizer: Apparently Jean Paoli’s team at Microsoft created XMLHttp in 1998 in order to give the Outlook team a way to do Outlook Web Access.

I mean, that’s not all that interesting. It’d be more neat if what we were using it for now was totally not what was expected, but I guess the tool is being used for what it was invented.

Full article about that, the Microsoft Atlas project, which I’m guessing is Visual Web Developer 2005 + ASP.NET 2.0, you can go to ScottGu’s Blog. There, he talks about the upcoming Atlas Client Script Framework, which will provide ajax support to ASP.NET

Next, we need to have System.Windows.Forms implemented in XHTML+CSS+AJAX. Wouldn’t that be something?

The even longer tail and the Adwords API

At first, the Google Adwords API doesn’t really seem like much, just a way to write simple programs to do some automating of tasks that are already pretty easy. I’ve bought ads on Google before, and it’s really fairly straightforward.

However, it’s really a MUCH bigger deal, in my opinion.

If you look back at posts like this one, you’ll see a nice discussion of how google is taking advantage of the whole long tail phenomenon by letting the massive amount of small guys with targeted keywords advertise to their respective markets. This is well and good, but there is something bigger that’s going on here. My speculation is that there is a much greater long tail than anyone has imagined, and that it can be served in a way that it can’t now — through creative use of the Adwords API.

The idea goes like this: There are lots of businesses out there that have a whole slew of products, services, and media to get to their respective consumers, but their focus is not targeted enough to make manually selecting keywords a viable option.

Take, for instance, a small business that sells a variety of, well, let’s just say ‘themed greeting cards.’ They’ve got their products available, but they have so many cards of various types that they don’t really have a good way of getting all their data into the adwords system.

Now, they can.

Anyone with a product catalog can do it. I’ll bet that in a year, good comercial ecommerce solutions will incorporate product catalog -> adwords (and the other guys) systems.

It’s going to get that easy to mass-micro advertise, and small businesses will benefit. So will Google.

Yes, there will be competition. I think Yahoo’s entry into this space will take a big dent out of Google, or at least it will put a little bit of pressure on bidding for keywords. Overall, however, growth in the market for keywords will outweigh any competition that comes online, at least for the forseeable future.

American Express RSS Feeds?

So, I was paying my AMEX bill yesterday and i noticed that they had an “RSS Feeds” Beta trial. At first, I was like, “Oh cool! They’re going to do something innovative with the technology!”

Then, I signed up for the service and realized it’s all just a big marketing ploy to get you to buy products that they’re offering.

To sign up for their marketing gimmick, you can go here.

Ajax Timeline

  • Late 1998 / Early 1999 – Microsoft releases IE 5.0 with support for XMLHttpRequest
  • Sometime In Between – Oddpost uses XMLHttpRequest to produce a highly functional webmail client
  • February 2005 – Google releases Google Maps to the world demonstrating cross-platform use of XMLHttpRequest (Google also released Google Groups, Google Suggest, and GMail using the technology around this time)
  • February 18, 2005 – Jesse James Garrett writes, “Ajax: A New Approach to Web Applications” on the Adaptive Path website.
  • The world jumps on the bandwagon, seeing how cool Google Maps is, and that it’s not just “google magic” — the name “AJAX” and some simple descriptions of the technology allow lots of people to really grasp what it’s about.
  • Everyone who is doing something like this already starts calling it AJAX, too. (except the google engineers, who apparently just call it “javascript” — how modest, and everyone who still called it “XMLHTTP”)
  • …
  • Profit

I’d like to point out that many of the “rich application frameworks” that are out there (many of which are described in my rich application frameworks page) have been at this for a while, working on their technology out of the limelight.

It’s just that several powerful trends have collided, and the naming of the beast has, well, given everyone a focusing point. Naming something like this gives everyone a common frame of reference. Not that “XMLHTTP” wasn’t a good name for the idea, it’s just that, well, it’s not as sexy as AJAX. The naming of the technology, the very prominent use of it in Google Maps, the already-breeding realm of rich application frameworks, well, all of them collided and produced this idea in everyone’s head that Javascript isn’t as bad as we had all thought, and that using “modern” javascript could really produce some highly functional, powerful web applications.

Part of it is that javascript left such a nasty taste in people’s mouths that it was relegated to the, “only use if it you have to” realm. I know that’s where I was, having beat myself over the head with the javascript stick back in the day when Netscape still had >50% marketshare. I think a lot of people are seeing this technology and are realizing that we’ve come a long way since then. Really, for the most part, now you *CAN* have one codepath (for most things), and you don’t have to hack, hack, and more hack your way to getting things working on various browsers. Part of it is that we don’t have to test on Netscape 3.0 anymore, and part of it is that the technology has matured enough to not give everyone headaches.

There’s more to this, for sure, but I think i’m hitting the key points.

Some of this was ‘researched’ at various places around the web, but i found the Wikipedia AJAX article very helpful

Bandwidth Use?

If you had massive amounts of bandwidth available in a datacenter, but for downstream-only, what could you use it for, commercially? Most providers of content have to buy large amounts of bandwidth, but it’s for outgoing traffic only. If there were a way to use the incoming bandwidth effectively, it would essentially be free.

You could provide backup service (where people backup over the net TO you), you could provide bandwidth to end-users (who download mostly), and, well, that’s all I can think of. What else is out there?