The danger I see is in conflating cause and effect. Is the App Store popular because the iPhone is great? Or is the iPhone great because the App Store is popular? There’s a big difference between those two arguments. The latter is the argument Microsoft has long made regarding the advantage of Windows: Windows is great because Windows has the most software and most developers.

It’s not one or the other, they are two different aspects of a computing platform. Looking at it like mac versus windows is helpful. The mac has super high quality apps for things people do a lot. Transmit is the best ftp app, Textmate is the best text editor (in my opinion of course!), Tweetie is the best twitter client. These are quality apps. The mac is great because it has the best software.

Windows on the other hand has more apps. But this is a honest feature separate and apart from quality. If you want to use a gui text editor you would be better off with TextMate on a mac, than Notepad++. If you want to karyotype some metaphases though you can either get Cytovision for Windows, or just not do it on a mac. There are a lot of niche apps that simply don’t exist on macs. Windows is great because it has the most software. If you’re using niche apps, you get stuck on whatever platform has the most software frequently. Sure, I’d love to use a well designed app for karyotyping, or for pathology informatics on a mac, but they don’t exist.

Right now the iPhone is great because it has the best software and the most software, but it’s not an either or situation.

I think parts of this are very true. Any plan that doesn’t count on software fracturing off into modules better suited to the user and business case is dangerous. Software is only going to get better, easier to make, and faster. The days of an immunology lab and a cytogenetics lab using the same software to do their work are over. The idea that they ever used the same software for their work is a conceit of vendors anyway, I think. They used the same software as a cash register, and did their work on paper most likely.

So the central point, that the LIS software environment will become more fractured and specialized I think is true, but the idea that the current LIS will serve as a viewer I think is giving too much credit to the current LIS. The current versions are not nearly as open and easy to integrate with as they need to be for this to happen.

EMRs are becoming this federation hub viewer already though, so do we really need another viewer? Especially one that is years behind? If I’m in the immuno lab, using special immuno software, do I need to see a federated view of lab info independent of the EMRs view? The case can definitely be made that such a federated system needs an LIS-side aggregation point for the data, but I don’t know how much user viewing functionality that needs to have. Most users would be happy to view one level up, at the EMR side I think.

Federation allows jobs to be done on the software most suited for the task. Many tasks are being lost to the central LIS (molecular, and other ancillary testing). This doesn’t mean that the main LIS isn’t still the best solution for some tasks though. Often reporting, and billing are its strong suits, and they may continue to be. I don’t think the future for “classic” LIS is in viewing things from disparate systems, that job is now the EMRs. The classic LIS needs to define and focus on its strengths though, (probably reporting and billing) and work to improve them, and their interoperability. If all the jobs get stripped away from the LIS they are in trouble. In a federated system if your app isn’t contributing then it’s not needed.

People outside of healthcare, I think, have the problem of thinking that since they infrequently interact with the system, the system is simple. Not everyone of course, but hopefully most people aren’t spending significant amounts of time in healthcare. That doesn’t mean healthcare goes away when you do though, it’s a huge industry. Every year over 2 trillion dollars are spent. 15% of US GDP.¹.

The calls to “computerize” need to be seriously considered. What do people mean when they say that? To take the example from the article linked above,

As a result, analyzing the effectiveness of specific treatments–for example, spinal-fusion surgery versus physical therapy for back pain caused by a herniated disc–is unnecessarily expensive and time consuming.

We could do research like this if we had all the healthcare data computerized in standard formats that were searchable. And I mean all the data, clinical, lab, imaging, survivorship, outcomes, (standards for measuring those outcomes!). Let me just say, this type of searching in a subset of results is one of my day job responsibilities, and even within one large hospital I would love it if this were true. The doctors would love it even more because they could publish even more, and make all sorts of discoveries.

While this level of computerization and standards would be fantastic, no industry is currently at this level of data modeling and standardization.

Oh sure, you say, look at google, look at microsoft! This is the misunderstanding. Healthcare is not a company, it’s an industry. Here are some fun mental examples of similiar technological problems,

• change a package from Fedex to UPS in mid shipment
• use whichever cellular company’s signal is the strongest wherever I am
• see how many jeans i purchased last year at Levi’s, Gap, and Express, taking into account the fact that Gap calls some of their jeans trousers incorrectly

The fact that I can’t do any of this doesn’t mean that these industries are “unmoved by the logic of lowering costs to increase profits”, or that they have failed to move to the modern era of computing. UPS and Fedex’s logistics and computing resources probably rival Google’s, but they certainly don’t speak the same language. Most retail stores use some crap command line POS (that’s point of sale, natch) system that looks like it’s from the 1980s, but it gets the job done. No one goes around lambasting them for their Luddite technology stance.

The examples in the article of electronic ticketing and stock certificates are ridiculous, so I wanted to make the examples a little more complex. (although not much, jean purchases aren’t quite the same as multi terabyte gene array data, or 4D real time diagnostic imaging, but…) The differences they have to a piece of paper saying that you can board plane 1477 at 7pm, or that you hold 100 shares of AAPL stock are hard to elaborate more on.

Don’t get me wrong, storing this data in better ways would be a dream for me, but lets not pretend that the healthcare industry is some laggard trailing every other industry in technology. Most industries don’t need interoperability, and they don’t have it. It’s certainly a laudable goal for healthcare, and I think that motivation can only appropriately come from the government. I would love it if the healthcare industry adopted CCR for everything and it was all standards based and interoperable, but the fact that we aren’t there now doesn’t mean healthcare hates technology, or is avoiding it to keep their costs high. I just want some realistic discussion that isn’t based on the IT situation at some tiny doctors office.

1. “Health Systems Resources” (Excel). World Health Statistics 2008: Global Health Indicators. World Health Organization. 2008. http://www.who.int/entity/whosis/whostat/4.xls. Retrieved on 2008-08-30. *Note: It is on the tab marked “4″ on the bottom.

The reaction so far has been very positive I think, but there has been some negativity. Most of it seems to be along the lines of, “that’s nothing new”, or “it’s just fancy chat” etc. I think a lot of this is missing the point from a perspective of technological superiority. Sure, it might not be anything that different if you are the kind of person who can set up a wiki, and a mailing list in a few minutes. Most people aren’t like that. Most people have never used a mailing list. Most people that use computers have never used a mailing list, or a wiki. It’s just too hard, and not something they are interested in. People that forget this I think, may just be (enviably) isolated from that level of user.

The inbox metaphor is something that people get. If I can send you a wave, that is also a wiki page, and a chat, and a shared document all in one that is a lot of hurdles I have crossed in getting you to use modern software. Currently I have to talk you into a wiki, train you to use email correctly. (no reply below what i said!) and convince you that chat isn’t just for your kids.

From what I have seen the initial user interface will be easy enough for everyone to just get, without having to be talked into it. That ease of use is based on the email metaphor. The actual protocol though is not similar to the email metaphor at all, and is another source of excitement.

This system is lightweight and simple enough for it to very easily grow into things that are not planned now. The system of waves, wavelets, and blips, allows for a really simple access control system, and versioning. As more clients and tools support this I think we will really start seeing some uses for the tech crowd that look very little like email.

I hope use of this spreads quickly. Microsoft seems like the big loser here. Of course they could implement this in whatever is after Exchange 2010, but somehow I think they would resist out of pride.

It’s a race to turn a relatively benign symbiote (the original Napster, which offered to pay for its downloads if it could get a license) into vicious, antibiotic resistant bacteria that’s dedicated to their destruction.

Exactly, from boingboing. Freenet is ready to go, people just need the motivation.

I think we might have lost track of the problem. The problem is that twitter limits your posts to 140 characters. Twitter does this because it started mainly as an SMS system, a stupid protocol that the mobile carriers thought up to make money overloading their already existing control channels. I don’t know what the break down in their traffic is, but I think we can safely assume most users don’t use SMS now. I just loaded the public timeline repeatedly to get a sample of 100 tweets, and none used SMS.

Twitter counts absolute characters also, so for example a tweet can’t do something like:

Hey, i love google!

And get a count of 19, they have to say:

Hey, i love http://google.com

And get 29. This is because, again they are sending to phones, and most SMS clients are stupid and can’t handle links at all. (which is usually smart since they don’t have much room to work with anyway)

To get around this problem, people turned to url shorteners. It took people a while to realize this was a terrible idea. urls should mean something, people like knowing where they are going, it destroys the link structure of the web, it relies on a company to exist and creates a single point of failure, and any number of other terrible things.

So the solution is rev="canonical", where I would say, for this page maybe

<link rev="canonical" href="http://sleepyhead.org/canonical" />

First, the philosophical problems. I like my actual url. It all means something, why change it? If the problem is that twitter can’t send links with it to mobile phone, so be it. The phones that people are using to read them via SMS probably don’t have a browser anyway, so who cares? Some people might say, oh but the 140 character limit is good for other reasons! It forces brevity. Great, count the displayed characters then, and let people put links like normal. No one complains that links are too long in html, because they are hidden. Since 99% of the platforms people use twitter on are fully capable of rendering html, use it.

For the real problems.

• http://sleepyhead.org/canonical is way too long. People use bit.ly more than tinyurl because bit.ly is shorter. If my short url isn’t as short as one you are going to use from a shortener, you won’t use it.
• twitter and services like that would have to ping the site to see if it has a short url, and then change the tweets that use the “wrong” short url? never happen

The real solution is to get twitter to:

• count the displayed characters
• allow people to add links like normal
• if someone pastes in a url, and it goes over 140, make it an embedded link on the word “link” or something
• don’t worry about sending links to phones via sms, the only phones that can handle this at all have twitter clients anyway.

I’m not really worried mind you, this has about as much chance of happening as most fancy link voting type schemes. (most of which I actually really like! haha) Which is to say, none.

How it works is pretty interesting. Topics are the big questions, “What laptop should I get?” etc. Under these topics are questions and results. Questions are the filters that lead to results. Things like “Should the monitor be bigger than 17″?” for example. Results are the solutions. There are two types of results, yes/no, and specific things. In the laptop example, a result might be a MacBook Air. Some questions, like “Should I learn to ski?” would have just two results, yes and no.

Results are “trained” which means for each result someone goes through all the questions in the context of that result and says which specific answers apply to it. For example, the topic “What font should I use?” would have a question, “Do you want serif, or sans-serif?”. The result Palatino would be trained in each question. For example, Palatino, would be trained as yes for that question. The training is done mostly by humans, but if the algorithm sees something it thinks correlates, it will train on its own. The ranking of results is based on a combination of the training, and what people like you have rated things. (people-like-you being based on the THAY questions)

It works surprisingly well really. Even without the training, and questions to guide it, it can do a pretty good job basing it’s answers only on people-like-you data. The team is also very responsive to suggestions and things in the forums, and is iterating a lot. It’s fun to see employees so obviously excited about the product they are making.

If I have one criticism, it’s that there isn’t enough to do, and no great ways to tell people what needs to be done. There’s also no good way to compare work and create any feedback loops. I think these things help build a real community. I’m thinking of this a little like Wikipedia, and maybe that isn’t a correct comparison. Maybe they don’t need a community in the same way really. It’s certainly very easy to use, and they have built the tools for people to correct their own results. Maybe that’s enough.

It’s a form system that stores the data in xml, for use by web applications, or more realistically Sharepoint Server. Businesses need to make forms, and Sharepoint only has limited support for this, so obviously the answer is another application to create forms, that will interface with Sharepoint. Because no one has thought or had this problem before, right? No one has ever needed to make a form on the web until Sharepoint came along, and helped everyone.

It’s just weird. Maybe this is a textbook case of escalation of commitment. The idea that you need to throw up some web form, and that infopath is the tool for you seems like a huge weird leap to me. You have to have it installed (even to fill a form out!), and it’s not free, you probably need Sharepoint, which is also not free, and functionally does everything much more poorly than, say, mediawiki, or knowledge tree. So instead of a simple server side client-agnostic approach, you would take the exact opposite, and sacrifice functionality?

Now for the meditation. People use this.

What if a browser displays a notification bar for something, and the user doesn’t notice it? Should it check to make sure?

Oh ie, so much fun.