The Validity of Internet News

Cringley has a piece from a week ago on Internet news. I do not usually like writing retorts but this one really stirred up a reaction, and since it was something I had not thought about much before it seemed fair game.

The basic premise of Cringley’s criticism of Internet news is that it is not timely enough. He mentions, but glances over the real time nature of some news, focusing upon the mainstream news sources. It is a fair tactic since it is true most internet readers do not search out those first person sources, and instead rely upon republished paper news.

But even here the argument is quite flawed. The problem is for some reason he assumes that a shorter half-life for news is important. The truth is that the vast majority of news is not news at all, and should have a half-life of zero seconds. On the internet, it is far easier to do that. In my opinion, for the average person, news that has a half-life of 24 hours is generally worthless. It really does not matter much to me that I hear about the AMD-ATI deal within 24 instead of 48 hours. The important news is the news that has long-term effects.

Sure, there are exceptions to this, but for those that need that real time news the internet offers better avenues then print as well. Besides there is TV, which is the best place for real time, filtered news. Print is locked into a single mindset, and that mindset just does not match up with the average person. Perhaps the 36 hours half-life of news on the internet has little to do with timeliness, and is truly a reflection of what users want. Print cannot supply a 36-hour half-life, so of course it does not have one. But maybe it should.

I am not sure how Cringley jumps from the statistic on half-lives to the conclusion that we read less, or the successive jump that reading less means we learn less. If you spend your day scanning through a newspaper, you are bound to read more, but did you actually learn anything? Anything valuable? A lot of this is bound to be fluff or sensational.

I can understand where his bias comes in. Cringley is a professional writer, for him it is all about gathering up arcane facts to make a compelling case for some audacious declaration, so suckers like me will feel compelled to respond. He needs masses of seemingly useless facts to do this.

For the average person, I think we would do better to spend a bit more time thinking about news, rather than constantly trying to “keep up to date.” It is a pointless game, and unless you are going to use this information in the next 24 hours, it really does not matter.

Where print and internet journalism really add value is through analysis, and in this arena the internet wins hands down as a source. Why? Because news does not decay like paper. When you are doing analysis, you do not limit your sources to the last 24 hours. That would be crazy. The internet offers a much better forum for criticism.

In truth, I cannot see any advantage to print journalism that stems directly from its distribution model. There are some advantages, but almost all of them are the result of social factors. Why don’t’ all of the New York Times stories make it to the Internet? Is it because the Internet is not big enough to hold them? No, it is because the Times is trying to hold on to its print customers. That is because they have had very little success selling online subscriptions. No one ever seems to consider that the people still paying for the print subscription might very well be the same kind of people who would pay for an online subscription. No, it is simply assumed that if these people had the option of getting that news online, they would convert to non-payers like the majority of the internet population.

Journalistic organizations will always exist. They are a very necessary part of the news cycle, but the idea that they need to be founded upon print publications is as wrongheaded as the idea that music must be based upon physical CDs. In both cases, the ancient organization would be best served by taking the plunge and providing something better on the new medium, rather than consistently delivering watered down versions of their bread and butter. Internet music should be higher than CD quality, not lower, and online new sites should have more information, and more timely information, not less.

Would I subscribe to such an online news site? Maybe not, but then again, I have never subscribed to a newspaper anyhow. But by not providing comparable or better coverage online, print publications are playing a dangerous game. If they provided as high quality or better coverage online today for the same price, many paper subscribers would switch. However, by not doing so they risk those users switching to an internet alternative that is as good as the watered down version. One thing any good sales manager will tell you is that acquiring a customer, or winning back a lost customer, is far more expensive than keeping an existing one. So when the day comes that the online concept is embraced, that former subscriber will be far more difficult to win back then they would have been to keep with a pre-emptive strategy.

Why we need a multi-tiered patent system.

Summary: A change to the patent system that provides short term and long term patents dependent upon the amount of effort and scrutiny applied by the applicant and reviewers.

In my last two posts (one, two), I talked about today's patent problems, and one idea to improve the situation. I said I had one more big idea. I actually have plenty of ideas, but this one I felt was unique, while others are commonly known ideas.

The details are important here, but at a high level, I think we need more than one level of patents. Today, there is one type of patent, the 20 year, exclusive and unrestricted patent. This may have worked well when all patents were of mechanical tools or machines, but today patents cover a much larger body of content. The one size fits all approach has driven up costs for acquiring, protecting and licensing patents. It has also left the door open for abuses by leaving the Patent and Trademark Office (PTO) with only two options: approve/deny.

To resolve this, I would suggest a multi-tier system. At the highest level, you would have a full patent grant similar or the same as today. There are opportunities to fix this tier as well, but rather than getting bogged down in discussing traditional reform theories I’ll plunge ahead.

Tier 1 - Automatic Documentation of Claims

The lowest tier, which I will call tier 1, would be very easy to get a grant for, but would have corresponding limited rights. I would imagine automated submission process could grant tier 1 patents, tied to an algorithmic search of the patent database. Knowing that this is a very low bar, a tier 1 patent would provide no value in enforcing claims, but would function more as matter of documentation. It would be useful to defend against claims from patent trolls, and would provide a stepping-stone to higher patent levels. In many ways, this is similar to filing a patent application today, with two big differences.

For one, a patent application is very expensive because it must fulfill not just basic documentation rights, but must also serve as a vehicle to a patent grant. Another difference is that there is no process today to request just the documentation rights, without seeking a general patent grant. These two differences leave inventors with two options: the expensive patent process, or secrecy. After obtaining a patent, it seems natural to the grantee to enforce it vigilantly, due to the rights and the costs associated with acquiring those rights. A tier 1 patent would protect an inventor from lawsuits, without granting rights that would encourage stifling competition.

Tier 2 - Fixed Public Domain Buyout

Of course, many or maybe most inventors will not find tier 1 patents sufficient, and so there would be additional tiers. The next level, tier 2, would be similar to a full patent grant, but with some key limitations. The most important is that tier 2 patents would have a set buyout price. The buyout price would be a set price any entity can pay to the inventor to convert the patent to the public domain. This ties in very well with my first idea, but would not necessarily need to be the government performing the buyout. It could also be individual corporations, industry groups, any organization or even a generous individual.

Tier 3 upto Final Tier with Current Law Rights

Additional tiers would fit between a full patent grant and tier 2, but they would be graduated variations of tier 2 that would vary the buyout price in correspondence to the amount of review, rigor, specificity and opportunity for contestation involved in the patent process. You might vary other factors such as durations in these tiers as well, but I think the most important factor would be the buyout price.

To save in legal fees, patentees would be encouraged not to always seek a full grant, but to seek a grant commensurate with the specificity and value of their claim. When applying for a full grant you would acquire the lower tier rights along the way, and even if denied at any higher level you would retain the lower tier rights.

It turns out that this idea is not entirely unique. What is unique is the idea of a public buyout price. I consider this a far better safety valve than limited durations. It is difficult to find a term that works equally well for innovation in fast paced industries like software, and in industries that rely upon lengthy manufacturing upgrades and implementations. Shorter durations increase risk for inventors, since they may fail to productize within a shortened duration. Buyout prices however give inventors a much clearer perspective on trade-offs. Buyout prices need not be trivial. Even at tier 2 I would suggest a buyout price of at least $250,000. Higher tiers could have buyout prices as high as hundreds of million dollars.

What do you think of this framework?

Why national governments should buy patents

Summary: By purchasing patents and selectively placing them in the public domain the government can both encourage innovation through post invention rewards, and remove barriers to competition and future innovation posed by restrictions on "core" technologies.

In my last post, I promised ideas on how to improve the patent system. I cannot say they will “fix” it, but I think they would help. I could start by suggesting that software patents themselves simply be abandoned, but this has been debated quite a bit and so far there has been no progress. My first idea is a bit more unorthodox, but I believe is better suited to making everyone happy, and in general encouraging society’s goals for patents.

The short form is, governments should invest in patents. It is not an entirely new idea (another example), but it has never been an institutional practice, which is what I would suggest. Also, while I think one goal should be to place patents within the public domain, I think there is room for true investment, where governments collect royalties upon patents. For several reasons this is necessary for the success of any large-scale program.

One reason is the international equation. If governments are not encouraged to invest in patents, but only to buy them out in the name of the public domain, there will be no reason for multiple governments to compete in patent acquisition. From the perspective of the inventors this is not a major problem, there is already plenty of competition in the private sector, but how would US/British/French/etc. taxpayers react if the discovered other governments of the world were not pulling their fair share. Encouraging investment simplifies this problem by encouraging the acquisition of patents directly, and leaving the concerns of release to the international public domain to the diplomats, legislatures and heads of states who love to argue about such details. In the end, we would hope for an international treaty establishing such an initiative.

Another reason is the concerns of favoritism that would arise from the government paying for an intangible like patents. I am sure we can all imagine the possibility of certain congressional representatives arranging “graft” patent purchases. But if most purchased patents are expected to perform monetarily for some period of time before being released into the public domain, then some measure of accountability is introduced.

The final reason is that a program that licensed some of the patents might have costs in the 9-figure range, rather than the 10-figure range. This makes the program more politically palatable.

Advantages

There are still some significant advantages to having the government own certain patents, even when it is collecting royalties on them, prior to releasing them to the public domain. One advantage is the government is far more likely to make certain concessions, such as giving educational and public works institutions royalty wavers.

As well, organizations may find it less difficult to negotiate cross-patent licensing agreements. The open-source community might find this to be a very attractive means to reduce their patent liability concerns. It is important however to insure such deals are not based upon favoritism. Open source organizations should be contributing to the overall patent body, not just consuming. This would mean licenses like BSD that contribute to the public domain would have no problem, but more restrictive tools like the GPL, that are based upon collective pools, would need to find ways to share those pools outside their closed domains.

Another advantage is that government issued licenses are more likely to have standard terms, rather than require case-by-case negotiations. This reduces risks based upon the unknown licensing costs when standard terms are not available, and saves on costs in general by not incurring legal and other costs associated with negotiations.

In addition to these short-term advantages, accelerated conversion to public domain would occur. I would suggest 3 years as a general rule, and much quicker for the most important inventions. Release would be based upon factors like the importance of a patent to future innovation. Profit patents would be those more focused on production, rather than bases for future innovation.

But which patents?

A difficult decision is which patents to purchase. An obvious starting point is patents that are necessary to future innovation and research. A good example of this is patents for research techniques in the biomedical industry. Because these techniques are useful to the design and testing of products, their broad availability is invaluable to future innovation.

It is tempting to suggest acquiring patents with the most value to society, but unless broad availability has a value greater than the acquisition cost, I would disagree. The point of this program would be to remove obstacles to future innovation, not to reduce manufacturing costs through government subsidy.

All in all, I believe a program like this could be a win for everyone involved. The inventors would have another possible buyer. The users of patents would have less difficulty obtaining rights, either through simpler terms, or through public domain release. Taxpayers would have some costs involved in supporting the program, but those costs should be returned through lower product development costs and faster innovation, producing better cheaper products.

I have one more big idea, somewhat complimentary to this one, which I will write about soon. I would like to hear what others think of this one however. I will not pretend to have spent years thinking about this, but ideas are like a ship. Most holes you can patch if you have not already dropped it in the water. I am sure I have some holes still, but this feels like a good idea once finished.

Why do we have patents?

Talk to just about anyone in the computer industry today, and they will agree patent law is flawed. Patents, in general, are not serving the purposes for which society supports them. Society supports patents for essentially three somewhat conflicting purposes.

1. To protect the value of inventors labor.
2. To encourage potential inventors to invent.
3. To encourage the free exchange of ideas.

The first goal is straightforward. Society supports this goal not just because any individual might be or become an inventor, and thus desire protection, but because in an indirect fashion it encourages the final two goals. Even today, this goal has some degree of success. Patent law itself is fairly explicit in protection. However, the real value of these protections is not as great as they seem. Schemers, trolls and other miscreants who exploit the patent system complicate the problem of enforcement. Legitimate claims fight an upstream battle while illegitimate claims exploit loopholes and extort settlement monies. In the realm of software, where patents are the most controversial, the value of legal patent protection is somewhat of a joke. It exists, but most companies acquire patents to protect themselves from illegitimate claims, rather than to defend a legitimate claim.

Whether or not patent law helps or hinders the second goal is more questionable. While there is some positive trickledown from protection, there are also a number of negative influences. If every invention were entirely unique, we would have nothing to worry about. However, not much is entirely unique these days. If a discovery or invention is not directly building upon the shoulders of others, it at the least incorporates some of the components of other inventions. Every one of these relationships becomes a liability to the inventor of today. The complexity, cost and risk involved in dealing with these liabilities is rising due to misuse of patent law. In the extreme overlapping rights create a situation in which an inventor finds it nearly impossible to put together the set of rights necessary to support an invention. This has been referred to as the The Tragedy of the AntiCommons.

The third goal should have been assisted by the open publication of patent papers. In theory, inventors could use this information to innovate further. If you have ever read a patent application however, you will know how far from this ideal we really are. Patent applications and grants are very often an undecipherable set of nearly meaningless high-level claims. At least that is my experience with software patents; it may not apply across all fields. In some cases, a technique called Submarine Patents has been used to entirely conceal all details for long periods.

None of these problems are new, so why am I writing about them? Well I have some ideas (another), and in the next couple of days, I will get them out here too.

Bad Data In, Bad Data Out

There is a phrase “Bad data in, bad data out”. It seems like an obvious concept, but it is not as perfectly true as it is catchy. It is difficult, but if you have enough bad data, you can create good data. So, if you want good data, there are two ways to get it. One way is to start with good data; the second is to have lots of data, and a smart way to process it.

This second way is at the core of successful “less is more” attempts. Sites like Digg or Reddit are great examples of this. Digg and Reddit do not rely upon the quality of articles submitted, they rely upon the vast number of users who rate interact with those articles in a semi-random manner, the vast amount of data they collect on those interactions, and a somewhat smart way of processing the data. They are not the best examples because they still have quite a bit of structure in their data collection, but they are a fair start.

Perhaps a better example would be search engines, like Google. These tools pickup a vast quantity of data in many different formats, with various levels of description, and use an very intelligent system to drag out scores based upon links, clicks, and lots of other random loosely structured data. The cornerstone of this example, is there is no built in, predefined concept of ranking in web pages. Most of the data that would be useful, such as meta-tags is notoriously “bad”. Still, those chaps at Google find a way, don’t they? The end result may not be perfect, but compared to the input, the output is not bad at all.

If your goal were perfection, you would start with a blank database and begin adding pages to it one at a time, setting a relevance rating for every term. The problem with this system is not that it would have bad data; it is that it would have incomplete data.

Believe it or not, I did not start this post thinking about data structures. No, I was thinking about corporate management practices. One fallacy that I see among management across a broad range of businesses is they often believe they can manage based upon a very limited set of data. The data they manage upon, financial data, is well scrubbed, well maintained, and generally extremely accurate. It has to be, because of regulations and because investors, suppliers and customers all want to insure they are not cheated. As accurate as it may be, it is an incomplete reflection of a corporation. As dirty as data such as customer interviews, employee discussions, theoretical technical designs, and innumerable other sources are, they provide a viewpoint unavailable from financial data alone.

I am not sure where this tendency is more severe, within upper-level management or within mid-level management. From outside, the decisions of the two are very hard to distinguish. Nevertheless, the reasons for decisions can be abundantly clear, even when their source is unknown.

When the workplace environment is methodically deteriorated, resulting in unhappiness and lost productivity, the reason is exceptionally obvious. Cost. It may not be cost effective, but it is easy to make it look that way. You simply ignore all that squishy messy data that tells you that people will not like this, that they will be distracted from work, etcetera. It is easy to ignore because it is “bad” data, which is hard to justify. The backlash can take a few years to show, the productivity losses will not show until the next product release, etcetera.

The adage I started with assumes something, it assumes that the algorithm behind processing is fixed, and does not adjust itself to the data. There is quite a bit of negativity these days toward anyone who “thinks with his gut”, as the phrase goes, the concept that only quantifiable facts are valid for decision-making is wrong. It is never prudent to ignore the facts, but when you know they are incomplete, and you do not have the time, resources or ability to fill them in, you are better off to take a guess than to pretend that the data you have is complete.

In a similar vein, delaying decisions to wait for better facts can be as risky as making a decision early. Ultimately, we have to use our intuitive abilities to make these decisions and guesses. Your intuitive ability is what varies the algorithm and allows using “bad” data to make good decisions.

How often we do this will depend upon what our job is. A research scientist, for example, should refrain from relying upon intuition to shape results or conclusions. Even here however, intuition can play a vital role in forming the original hypothesis, and in spotting problems with test validity. How often have scientists discovered the errors of their tests because they “felt” something was wrong? In this case, intuition is followed by experimentation, quantification, and validation.

In other fields, such as management, this is not an option. When faced with the choice of whether to upgrade the aging server hardware, commissioning a study to evaluate the effectiveness of upgrading that hardware may sound like a reasonable way out, but if it costs as much or more than the hardware, then you are better off just upgrading the hardware.

You have to know when is the right time for more investigation and when to act and monitor. Most professions do not yet have a guideline for this, and probably the only ones that ever will are those that are very much towards the “no acceptable error” side. It may be tempting to treat all professions like science, but science is expensive, and takes time to develop. Likewise with good data. Use it when you have it, but don’t forget about the data you don’t have.

The Future of Software

Note: I originally wrote this post in February 2006 for my “corporate” blog, but since its very general in nature, I can’t see any reason to keep it there.

The Last 20

Here is a question to spark some discussion: What is the number one thing that will differentiate software 10 years from now from software today.

Over the last ten years, I say communications is number one. Though the Internet existed long before, this time was the fulcrum point (I'm fudging a bit to say 10 years but who wants to be exact). A secondary and connected area is data. Data is secondary because without communications none of the vast amounts of data currenlty available would have been collected or widely available. In fact, most of the data that fills up the vast coffers of the World Wide Web was around prior to it's inception, but was never so widely available.

The 10 years prior to that, I say user interfaces. Computers during this period evolved from being designed for "geeks" or narrowly trained workers (data entry workers). Users could be trained to use a single program very well, but once outside that area of specialized training there was little cross applicability. Developers generally designed software that "they" wanted and could use, and did not design software for users. Software came with thousand-plus page manuals and developers expected users to read them (at least once).

Eventually, developers began to realize that if software was easier to use more users would use it. Also, users would have the time (and memory power) to learn several different pieces of software. One way software became easier to use was through convergence. User interfaces started to share common elements, metaphors and more, and not just within a specific silo (like word processing), but across many areas. Also, things like online and context sensitive help encouraged users to "learn on the job". Lastly, there was graphics. Mice, pictures, graphics, all seemed more natural to users, and applications could use visual tricks well known from newsprint and magazine trades to draw users to important details.

The Next 10

Ten years from now, I think people will be saying the use of statistics. Over the past 10 years a huge amount of information has been collected and everyday more and more sources of data appear. RFID promises to create a vast abundance of data on things industrial, commercial and even personal. Wireless technologies promise to make this data available wherever you go, and also provide a path for more real time data to be collected.

So after talking so much about data, why do I think statistics will be number one? Well because honestly there is too much data already. Statistics however are what allow this data to become useful. I know, I know, this is nothing new. Search sites have long since been using statistics to allow users to dig through the vast pile of data. We internally have been using statistics to provide data to our clients. But the difference between today and ten years from now is that today such applications are few and generally specialized. In the future statistics will become an integral part of almost all applications, and will apply to a much greater number of a user's interactions with software.

The trends are already beginning. Amazon.com was an early innovator, and a great number of similar or connected sites are taking similar routes to simplifying a users search for what he wants by comparing his tastes with users of similar tastes. Estimating of many kinds (think projects, inventory control, etc., not just automotive) will become based upon statistics rather than heuristics, which will increase its accuracy and cut administrative costs dramatically.

So what does this mean for us? For one, developers who do not already work with statistics would be well advised to begin brushing up on their understanding of statistics. While it is likely that the process of implementing statistical models in software will greatly simplify, an understanding of the core concepts is always useful. Database software packages are where most statistics related development occurs today, but I expect a number of specialized tools will spring up as databases struggle to meet evolving statistical needs while still delivering on their core principles.

As I close, one cautionary note about using statistics. Statistics are a tool like all other data. People still need to rule, not numbers. If you don't understand what I mean, think of how unpopular Amazon.com would be if instead of suggesting books, CD or movies it thought you liked it automatically mailed and billed you for them... The website can save you time because it might take you much longer to find something you liked, but once you see it you are far better off making the decision on your own.

An Inconvenient Probability

Earlier this week, I saw “An Inconvenient Truth”. In case you have not heard of the movie, it is a documentary on global warming, featuring Al Gore. There were parts of the movie that I really respected, and parts I did not. It is unfortunate, but I get the feeling that many theories that would have been better labeled as “maybe”, were not, for dramatic effect.

For me, I see global warming as a real threat that we should take seriously. This is in spite of not yet having seen anything that 100% conclusively proves the theory. 100% is a high standard, and when the down side is global catastrophe, action is justifiable at a much lower standard, and a few very basic facts can establish that standard.

One fact is that humans are changing the atmosphere by contributing additional CO2 gas. We may not absolutely know what the effect of additional CO2 gas is (I’ll get to this), but we do know we are producing it. You can measure the amount of CO2 emitted from a cars tail pipe, a power plants smoke stack, etc. The production of CO2 gas is a measurable, identifiable and provable fact that even I think everyone should be able to agree on without much controversy.

In 2002, worldwide 32 billion metric tons of CO2 were emitted. The atmosphere itself, has a mass around 5,000 trillion metric tons. By mass, this is a contribution of 6.4 ppm (parts-per-million). Ppm is usually measured by volume and at atmospheric conditions CO2 is 1.5 times denser than air, so we reduce that to 4.3 ppm by volume. Atmospheric sampling, has shown that the actual concentration of CO2 in the atmosphere has been raising by about 1.5ppm since 1978. Sure, the environment has natural fluctuations of its own, but if we are producing 3 times the observed fluctuation, it is likely we are having an effect.

We also have data that shows over long periods of time changes in CO2 concentration and average temperature have occurred at the same times. This data does not prove that CO2 causes global temperature increases (though there are plenty of valid theories that suggest it). What it does prove, almost beyond any possible doubt, is that the two elements are related. It is possible that temperature changes caused the increased CO2 levels.

You could create a theory stating that when temperatures rise, vegetation suffers, thus causing less CO2 conversion, and allowing global CO2 levels to rise. If you, and others, could not disprove this theory, you would have a second option.

Even if you did have an alternative theory, that you could not disprove, that would not disprove the global warming theory, which has stood up to a great deal of scrutiny. Unless I missed it, no one has provided any data disproving the theory of global warming, skepticism has focused on the lack of absolute proof. With a question as big as the climate of the planet, an absolute proof is currently beyond our means.

Without disproof, you are left with a significant probability the theory is true, and since the downside is a smaller car, and the upside is saving the human race from extinction, it feels like we should be doing something about it. If someone proves that global warming, and all other negative effects of pollution, are not threats, then I will reconsider. “Skepticism” seems woefully inadequate to justify our current actions.

The movie actually makes this point, but it jumps past it too quickly, and goes on presenting possibilities as certainties. This is unfortunate, because while it may initially feel more persuasive to present a unified front of certainty, it opens you to attacks from the skeptics. The skeptics distract from the real question of probability and the tradeoffs, with controversy, and alternative theories, for which they are no more able to prove, and are far less tested.

Multi\Dual Monitor Remote Desktop Goodness

One of my pet peeves with Remote Desktop has been that it did not support multiple monitors. Well in a sense, it does now. Vista includes a new version of the RDP client, version 6.0, which removes at least most of the limitations upon display size.

The trick can be found here. To sum it up a bit more concisely, plus a few added tips of my own.

• Create a folder for you RDP 6.0 files (don’t place them in System32)
  
• Copy the files mstsc.exe and mstscax.dll from a Vista installs System32 folder to the new folder.
  
• Create a subfolder, en-US.
  
• Copy the files mstsc.exe.mui and mstscax.dll.mui from the Vista installs System32\en-US folder to the subfolder.
  
• Create a shortcut to mstsc.exe and include the command line option “/span”.
  

This will get you the new client. You will still need to manually edit your RDP file to set the resolution you want. Normally you’ll find a default.rdp file in your My Documents folder, which is easily editable with Notepad.

It still does not support actual separate desktops (meaning that all Windows will maximize to both monitors, rather than just one), so you need to grab a third party tool for that, which as of yet I have not found. Anyone have a suggestion? I am pretty sure I read about one a couple months back, but not having solved the first problem, didn’t really take enough note, and now I can’t find it.

While I was tinkering with the RDP files, I ran into this little setting “remoteapplicationmode”, which lead me to a post about new features to remote individual applications without a desktop, which I guess will be part of Longhorn Server.

One last tip, I end with is to remind you that if you can redirect the handling of the open and edit commands to the new application. Unfortunately you have to do it through the registry because if you select the mstsc.exe in your new folder, Windows continues to run the one in its System32 for some odd reason.

Update: Microsoft has released this update for XP as a "Optional" update in Windows Update.

Update #2: Windows 7/Server 2010 will feature an even better version of this feature.

Web Interoperability and Identity

I read an article recently that touched on a two topics I have been thinking about writing about for a while. I had not really related to two topics, as I was coming at them from different angles. The first topic is the incompatibility of different but similar web-based applications. The author is mostly concerned with the proliferation of user accounts, but I was looking at it more from the perspective of how the standards mantra of web development is not keeping pace with the rapid development of new services.

A lot of smart people have assumed that web applications are magically protected from the isolationism that they criticize desktop applications of. It is easy to see where this misconception comes from, today on average web applications have a higher success rate in interoperability, and it was higher in the past. With every passing day however this success rate falls.

Web based applications do have some natural advantages to interoperability, but they also have some significant and often overlooked obstacles as well. In terms of advantages, there is the heterogeneity of the data stored online, and the providers’ ability to manage and manipulate this data to enable interoperability. That ability is provided by not having to worry about the connection quality from a desktop, not having to worry about deploying updates to interoperability logic, and being able to observe, troubleshoot and correct errors when they occur.

But at the same time, the fact that the provider has the data presents a potentially huge obstacle to interoperability if the provider is non-cooperative. With a desktop application, if the provider is non-cooperative, you can hack the format. That can be a lot of work, but it is certainly much easier and less likely to get you sued then it would be to hack a web provider.

In truth, I believe that the desktop application platform long term has better merits for interoperability, because the advantages of the web platform are being addressed. Broadband connections are very reliable, and once Wireless goes through what will most likely be a tumultuous start, there will be no big “reach” jumps needed to sacrifice reliability for. Automatic update services are becoming a standard application feature, and gaining more standardized tool support. Lastly, error-reporting services are gaining more acceptance (if Google knows everything you do, what’s the problem with Microsoft knowing when you crash?).

The web interoperability disadvantage, however, is not going away it is getting worse. As the number of web applications explodes, and the number of providers per service reaches ever higher levels (ever check how many del.ico.us copies there are?) interoperability becomes unprofitable to providers. Another contributing factor is that more and more companies with not as altruistic intentions are adopting web platforms.

Even if standards can catch up with development, they may find that adoption does not happen as easily as it has in the past, as providers refuse to share “their” data.

Coming back to the article, the second topic was about treating users as users, and separating their identities from the different companies they do work for. Separating their identity offers the opportunity to unify their work environment, while still maintaining the individual and company based controls that both sides need. It is not really a one to one correspondence, but I’ve been thinking about how to treat different companies in connected systems, and how to insure each company their own space. Choosing the right place for borders is important to the establishment of the agent concept. I completely agree with his concept on how to setup user accounts (he forgot to mention having a personal user space, but I have a feeling he was thinking it).

Reviewing Documents

Considering how much work goes into reviewing documents, designs, books, code, etc., I find it surprising how little guidance there is on the topic.

Quick Google searches pull up mostly information technical features of programs, like Word or Acrobat, to compare documents. Digging further brings up some scant information on holding reviews, mostly focusing on planning aspects, and sometimes giving useful advice for authors, like “listen to make your ideas better, not learn to defend your ideas.” What is totally lacking, however, is any advice on how to participate in a review effectively as a reviewer.

This is interesting because the way reviewers participate can have significant positive OR negative impact on a review. The worst reviews gather a large group of people, none of which have read the documents, and spend inordinate amounts of time focusing on nit picking little details. Good reviews use many communication channels, separate reading and discussion times, and only use discussion time for topics that:

1. Require group input or consensus.
   
2. Are intended to convey meaning to the group.
   

To a degree, I believe the reason why advice on reviewing documents is neglected, is people assume it is the author/facilitator’s responsibility to insure quality reviews. In part, this is true, but there are limits to the influence of authors/facilitators.

So how can you do better as a reviewer? The number one task is to read the document. Sure, it sounds obvious, but it is so commonly not done, that it is worth the reminder. If you do not have time to get it done before the review meeting, consider selecting a few sections and focusing on those. It might be best to try to choose something other than the first sections, since others may start and end here.

We all know time is limited so I doubt everyone will comply 100% with this suggestion. When I have failed to read a document prior to the review, sometimes I have been left wondering if I would have been more effective to skip the meeting and spend that time at my desk reviewing the document. This is especially true if no one else had read the document, and thus no topics of real importance were discussed.

My second tip is to choose the right channel to communicate. The worst thing you can do as a reviewer is monopolize discussion time to convey grammatical issues, style issues, and other one-sided issues or nit-picking issues. Depending upon the type of document, and its purpose, this kind of information can be extremely valuable, but it does not warrant discussion.

The best thing to do with such issues is to send them to the author before the review. Usually authors, when they get this kind of feedback, will simply say “doh!”, and fix the problem. If you and the author agree, there is likely no need for discussion. If you disagree, let the author ask the question. On highly opinionated, but relatively unimportant concerns like style, it is usually best to throw your 2 cents in and drop it there. If the author likes the suggestion he will say “Why didn’t think of that?” If not nine times out of ten, you will not get your way, and even if you do, there may be little benefit.

In some cases, like code reviews, it may make sense to have the author communicate which suggestions he accepted, rejected or feels require further discussion. More traditional prose documents, for issues like grammar and style, fit a fire-and-forget style of suggestion better. If you make the process to heavy for you or the author, either you will avoid making suggestions that could be valuable, they will unduly burden the author, or both.

The way I do this, is I take a copy of the document, and revise it the same way I would a draft of a document of my own (It’s important to remember to use a copy, or you’ll receive great, and justifiable, ire from the author ;p). You should use a tool that has revision tracking, and if possible, make sure that the only revisions that show in your copy are your own. Also, find another way of indicating the big issues, like as a separate email, or only use comments for big issues.

One of two problems I have with this strategy is that I can end up too focused upon the minor myself, and not see the bigger issues. On the other hand, that first read turns out to be fairly thorough, and if I go back and skim through again, I have a much better understanding.

Spotting the big issues is the most important part however, especially in early drafts. Quite often, you’ll just produce more questions. You might have additional content that needs to be covered. Whatever it is, big questions will usually require a second go around or more. These are the types of issues that might warrant discussion.

As important as it is to use the correct channel for small issues, it is equally important to insure that authors are given forewarning of big issues. It is impossible to do this if you did not read the document ahead of time, so this is one more reason to make sure you read early, preferably the day/night you get the draft. One reason why this is so important is so authors can be properly prepared for a discussion, and will not be forced to respond with a “I’ll look into that…”. Another reason is to help the author maintain his cool, and not get defensive. It is, at least in my opinion, a lot easier to take objectively criticism if it is not piled on in large doses in a 1-hour discussion. It is also far easier to take if when you are not put on the spot and (figuratively) asked, “Hey you, did you screw up?” If the author can respond, “Yes, I made a mistake, but I know how to fix it”, not only will more get accomplished, but he will feel better about the experience.

The second problem I have with my strategy, is that it is important to insure you are changes are properly received. “Rewriting” a document can be interpreted as presuming too much. In truth, it is the most effective way to communicate suggestions, but it is quite easy to see how the receiver might interpret it as assuming your way is right. However, as long as the author knows that he can either use or ignore the suggestions, you should be okay. I think it is better to offer an alternative, than to simply plug in a generic “confusing, try rewording?” comment.

Of course, all this comes from the perspective of a programmer\architect. I wonder if what works is fairly universal, or if it’s a lot different for things like books, movies scripts, marketing plans. What do you think?

The Virtues of Dual Monitors

Do you have enough desk space for this? While we may not all need one of these at the low price of $5,999, I am still amazed at how slow most technology companies are upon figuring out the virtues of dual monitors setups with cheap, affordable $200 19” LCD’s. There are plenty of valid studies showing significant performance gains from dual monitors in comparison to single monitors, and when you compare that $200 to an employee’s salary it requires an incredibly small percentage gain to be a good value. If you figured an employee cost at $100,000 per year (which is probably low once benefits, taxes and such are added), you need a 0.2% productivity gain to justify the cost. And that’s assuming you need to buy a new monitor ever year!

If you work with computers for a living, even if it’s not programming, you should probably consider this advice. There is a high degree of probability your computer already has the video card necessary. If it does, setting up two monitors is about as easy as it gets.

The JavaScript VM?

There’s an interesting trend going on these days, where many languages are adding features to compile their code into Javascript. For example Ruby, Java, C# are all doing it. Supposedly, the original idea came from a company called Morfik, but I know very little about that.

What I find interesting about this is that JavaScript is essentially being abandoned as a language and converted into an IL. Not having much respect for JavaScript myself, it feels very much like a first class hack, and I wonder how long it will live on. It’s easy to see why it is happening, JavaScript is already everywhere, so why not repurpose it?

But thinking longer term, clearly there must be better alternatives (CLR and JVM) than JavaScript.

Statistics don’t lie?

A long time ago, I posted about why HD-DVD was, in the long run going to beat out Blu-Ray. Well it appears as if it is possible my theory will never be tested, because it is based upon the concept that Blu-Ray would actually offer higher quality, even if marginally so, than HD-DVD. Surprisingly though, the word from users who are testing both, is rather bad.

Of course, there is a survey that claims 58 percent of consumers prefer Blu-Ray. But that survey is completely bogus, not just because of the obvious bias of those conducting it, but because parts of the results just plain do not make sense. For example, the survey claims that one of the reasons Blu-Ray was preferred was because of better backwards compatibility and compatibility with computers?

There are several problems with these comments. First off is that Blu-Ray does not have better backwards compatibility, or computer compatibility. Quite the opposite. HD-DVD is the format that allows a double sided disk to have one side in standard DVD format and one side in HD-DVD, meaning that you can buy an HD-DVD disk like this today for your DVD player, and buy the HD-DVD player later. In terms of computers, Microsoft plans to include HD-DVD support standard with Windows Vista (the exact level of this support I am unsure of, but whatever it is, it is more than Blu-Ray is getting).

The second problem is besides these being totally contrary to fact, they are also totally contrary to what you’d expect an uninformed consumer to think. If you had zero facts, and someone asked you whether HD-DVD or Blu-Ray was more compatible with DVD, which one would you guess? Duh, the one with DVD in the name. The press has also been reporting the stories in the same way. So basically there is only one way a surveyed would get the impression that Blu-Ray was more compatible. Yes, you guessed it.. the surveyor.

Nice try Sony, but at least make your lies believable.

Marketing & Life

Today I stumbled upon a series of articles about marketing, and I liked them quite a bit. What the author says about marketing practices is quite harsh, but true. It is not that marketing people are evil, but the rules of marketing generally encourage it. At one point in history, marketing may have been more about “getting the word out”. Today however, that is really not the case. There may be a little of that left, but in a consistent series of one-upmanship marketers have layered many other undesirable behaviors on top.

It is easy to blame the marketers for this, but the truth is, it is a game that is naturally setup to take on this shape. From the perspective of CEOs, marketing has one measurement that matters: how much product is sold. When judged this way, anything goes, and the only consumer benefit of marketing, information, takes a back seat to the manipulative tricks. This has gone on so long, that there many practices are simply accepted.

The result is very ugly, but effective, ideas like BusRadio. Things like this are clearly not in the best interest of society.

The author has a lot of good ideas, but I think the number one is the suggestion that no one is totally immune to marketing. I would consider myself relatively immune to marketing. I judge this based upon what I buy and do, rather than attempting objectively judge whether marketing has had an impact upon me. It may sound arrogant, but in my admittedly biased opinion, I am doing far better in this area than most people. But, even I can see some obvious signs of the effects.

In my case, I feel that the source of these effects is not (mostly) from direct advertising, but from derivative effects. See, even if you are immune, or relatively so, to the subtle and trashy tactics of marketing you still must contend with the influences from the social effects. By shaping the society in the large, marketers have three other tools available.

Tools

Word of mouth is a powerful tool because it is far more difficult to discount the opinion of friends and coworkers. Word of mouth covers not just recommendations, but the subconscious group mentality, and competitive nature of people. Even if you think your neighbor is a jerk, and hate that huge SUV he bought, the effect of him buying it may encourage you to buy a more expensive car than you would have otherwise.

Peer-pressure is another tool. Unlike, word of mouth, where you suffer not at all if you avoid the trap, peer-pressure is more sinister and active. The best example of this, is familiar to anyone, beer. How did it come to be that people will actually look down upon people who don’t drink beer. In all honesty, it is usually a pretty minor thing, but it is amazingly pervasive, and there are even a few zealots of this extremely minor and unimportant quality. The point is that you can actually suffer from not caving in to peer-pressure. Quite often peer-pressure and word of mouth come as a package, and you are left wondering how they are mixed.

The final tool is unavailability. When social norms shift too far in one direction, you may actually find it impossible or exceptionally inconvenient to not follow the crowd. The prevalence of cars in almost every American city is one such example. How many American cities can you even just commute to work and back conveniently through public transit? Shockingly few, and even in Chicago, which has either the second or third best US public transit system, it’s difficult not having a car, even if you don’t use it every day. The SUV craze is similar. Because so many huge and dangerous SUVs are on the roads today, others are forced to buy bigger cars or SUVs to be safe. If your sub-compact is hit by a mid-size car with excellent safety, you might be ok. If your hit by a 5,600 lb. Armada, your chances are slim.

What is important

I do not totally agree with the common views of what is important, sometimes the author slips into some of the common preconceptions. Some of these preconceptions are a reflection of the very things he’s talking about. I don’t believe everyone’s views on this will be the same, and I don’t think they should be, or that we should suggest they should be.

Take “culture” as an example. It is a common conception that this is something everyone needs. But this is in strong opposition to his views on movies, TV, etc. Sure you can view these as a lower quality versions of more traditional outlets of culture, but at the root they all derive from the same source. Ultimately the real question is whether you enjoy the process.

Another example is work, which common conception demonizes. Work doesn’t have to be this way. I would say that this conception adds just as much or more stress to people’s lives, as the material way of living. In another sense, they are symbiotic evils that feed of each other. If you go to work afraid to lose your job because you are in debt, you are much more likely to hate it because of the added stress. Then you might come to the mistaken conclusion that the way to fix this is to get away from work as quick as possible, and fill up the rest of your life with lots of meaningless filler. And is this filler free? No, it certainly is not.

So what would happen if you resisted the material treadmill? Well you likely wouldn’t have as many debts. There would be a certain point where you would say, enough is enough, and what I have is enough.

From here, one of two things could happen. One is you might work less, and the other is you might start to like working. Working less is not the end all solution. For one, if the goal is just to work less, how much less is enough? Will you never be happy until you are not working at all? What if no one was working?

I would much rather see people begin to enjoy work, to the point that they will do it not just to pay off their unnecessary purchases, but because they realize it is critical to society.

What about families? Children are important, but raising a child is another form of work. The same basic principles apply here as well. You should like it.

Solutions

One thing I didn’t like about all of these articles, is there only solution offered was, “Be informed”. While I think that is a great personal strategy, it is unlikely to change society. I think we need to do more than that. Unfortunately, I can’t think of anything more creative than government regulation. That of course is quite a can of worms, but if it’s necessary, should it be considered? Somehow you have to change the rules of the game such that it stops feeding upon itself to reach ever higher levels of depravity.

Could we set limits upon advertising in a reasonable, administrable way? Big question. I’ll think about it, maybe you can too.

VW High-Precision Auto-Navigation

I found this somewhat interesting. It's unfortunately overhyping trivial qualities like speed on predetermined courses, but the precision of the sat-nav system is a very important quality for a transportation concept like I'm thinking of. The whole speed thing is silly because in the real world, auto navigating cars won't be F1 racing. They may hit high speeds, but that will be in controlled, easy to maneuver environments like highways. The hard part is the unpredictables.

It's interesting to me because it ties in a little bit with my transportation concept.

read more digg story