actual progress is being made

I am in communication with UPS, USPS, FedEx, and DHL in regards to re-selling their online services, bundled and branded together in a single interface. FedEx's registration even asked which other carriers' services would be offered along with theirs, so they have obviously worked with ISV's that build shipping management solutions before.

mine has a few key differentiating factors, which I think will make it a great service.

1. built on top of a solid SOA. this is still in progress, but with a proper SOA, keeping up with all the different carriers' online functionality should not be a problem, and developers can be assigned to certain carriers, and to implement the carriers' functionality into the larger shipping management product that we are building. In addition, SOA based on WS-* standards will allow our offering to be incorporated into BPEL processes, which many larger companies will undoubtedly be using in the future.

2. low cost. we use open-source technologies, which makes our software costs approx. 0. I'm almost positive, should demand go as high as I'd like it to, to deploy on Zend Platform, or ActiveGrid, and to use commercial licenses of MySQL, Linux, etc. besides just the software licensing, the popularity of the OS technologies means there will be a huge number of developers as potential employees.

3. relating to a project Matt has going, I will also work on a RAD framework with him that will, hopefully, be built with consideration of allowing developers to quickly and easily set up communications with services such as the one we'll be offering. if that really is the case, then popularity of that framework will only help to accellerate the demand for our service.

4. federated identity. I don't know a LOT about this (yet), but the concept sounds great, and it's one of the differentiating factor that the Rearden group uses. hopefully Rearden will eventually be one of our customers. =)

IBM method for SOA'ing yourself

IBM got this cheer-leading help at internetnews.com, which is fine, because they're correct about lots of things. IBM wants to sell "SOA assessment services" just like I'd like to sell. well, they're calling it SOMA, and I'm calling mine, uh....advice?

in any case, with IBM's new partnership with Zend, it would be interesting to find out if IBM would ever be offering these assessment services to small businesses, and if they might pitch the Zend Core (including PHP) as an SOA platform. I, of course, think it's possible, but un-proven. and will hopefully move it into the "proven" column for my own sake, and it wouldn't hurt Zend and/or IBM to follow my lead! ;) man, that would be sweet.

however, I would hope that Zend and/or IBM is as open to an SOA platform that uses MySQL as opposed to the Cloudscape database in Zend Core. don't get me wrong, I think Cloudscape is a great project and I really like that IBM is promoting it. I just have a soft spot for MySQL.

repeat article?

I may have linked to this article, or a copy of it, before. I think it was called "The Killer Web Services App". this time it's touted under an SOA banner.

in any case, if I didn't already say so, I'd like to change the label from "The Killer...App" to "A Killer...App." I'm with Bezos when he says that we are only just seeing the beginning of these kinds of substantial new software programs that utilize the internet for base functionality and expand upon that base functionality for their app, which can be utilized in turn.

what I'm interested in finding out is what kind of platforms that killer app runs on. I can imagine that there is a case for a similar business that would be based on a LAMP stack and could thereby keep costs lower, and thereby offer similar or better services at a lower price.

also, I'm pretty sure I need to figure out a way to run said business while playing video games all day.

visa does a lot of transactions

proof.

when I first started reading, I thought it would be insane if those thousands of transactions per second were web services transactions, because those would be some pretty fat message stacks and I wouldn't even want to imagine the infrastructure required to keep them going at that pace.

in any case, the dispute-resolution scenario seems very conducive to a web services approach. and it made me think of the WS standards in a different perspective - outsourcing. essentially, Visa, member banks, and everyone else that uses the various WS standards is outsourcing the development of their internet-capable communication infrastructure.

I thought of it this way...

without WS standards, if Visa were to accomplish the same thing, they would need to spend time and money developing a method by which all of their member banks could communicate, via the internet, to their backend system for the dispute-resolution process. however, with WS standards, they can settle on the publicly-available HTTP, SOAP, REST, FTP, etc standards.

the article did not go so far as to say whether or not they were using things like BPEL4WS to model/manage the business processes involved in the entire procedure of dispute-resolution, but if they do, then that is yet another methodoly/platform that they did not have to develop in-house.

obviously, for those companies that pay to help develop these standards, it's not entirely free. however, it does help the standards-developers stay focused on their standards, rather than having to worry about this or that implementation of the standard.

that is all for now, though I'm still reading a few more articles, so another post may yet happen.