Fractals of Change

The Best is Yet to Come

Imagine that you had a magic parts bin with an infinite supply of instantly available free components for almost every function; just think what you'd be able to build! That nirvana is already available to those of us who "build" software. Computing power is cheap and getting cheaper; the price of data storage is coming down; and software "components" – scripts, subroutines, code packages, libraries, open source operating systems – are free.

As a programmer, I'm Rip van Winkle; didn't write code for seventeen years and now I've started again. All the young hotshots who were just learning to eat solid food when I last programmed take all this free code for granted; to me it's a wonder. Even at my age, I can build much more interesting stuff in a day of coding now than I could in a week the last time I was doing this – all because most of the functionality I need already exists for free picking off the web. I just have to find it (usually with Google), download it, understand it, lash it together, add a little secret sauce (which I promptly make open source), debug to alpha, and voila.

It's not that us ancient programmers didn't understand the value of code libraries and reusable code. When I ran a software company, I insisted that we componentize and build a subroutine library to make the next project easier. We negotiated contracts with our customers which allowed us to retain ownership of code that wasn't product specific so we could reuse it. We even kept the software library as an asset on our books – albeit a fast-depreciating one.

We got a competitive advantage from reusable code; so did anyone else who succeeded in the software business. Libraries were a huge asset when I was at Microsoft – and one that Microsoft had every right to.

Developers of new machines and new operating systems learned that making rich code libraries available to developers meant more applications would get written for their platforms. The Macintosh would never have succeeded if there hadn't been vast quantities of sample code to speed and guide our development for this new environment. BTW, most of this code came on floppies; some could be downloaded over our painful dialup connections.

I'm sure companies still have their own libraries of reusable code and benefit from them.

But now all the private libraries are dwarfed by the cornucopia of free code on the web. The first beneficiaries of this largesse are we programmers. Rather than diminish the need for programming, it makes each one of us much more valuable in terms of what we can produce – and produce very quickly. The ultimate beneficiaries are the computing public – which is anyone with a watch, a cell phone, a DVD, an modern appliance, or a computer.

Reusability of code is a network phenomenon: its value grows with the square of the number of people who are contributing and sharing. We've only begun to reap the benefit of global code sharing. The best is yet to come!

Vista and Internet Explorer: Ugh

The good news is that Internet Explorer in Vista has this feature called "Protected Mode" which "should" provide good protection from things that go bump in the night on the Internet. The bad news is that it mostly protects you from yourself. Most people I know turn protected mode off although it does seem to be working fine for Mary. It definitely doesn't let me do active-x installs even though I've turned on that option everywhere I can find it and I run as the administrator.

Worse news is that, protected mode or not, IE just seems to crash or lose function after a while. It's the same IE that runs under Windows XP but it's not nearly as reliable in this environment. Some Google pages don't work – some of the time; other sites randomly fail. Some flash stuff won't image – sometime. Some YouTube won't work – sometime. And worst of all, stuff I'm testing sometimes works and sometimes doesn't.

Restarting Explorer sometimes – usually – solves problems. Rebooting the whole machine makes other problems go away. Browsing with Firefox rather than Explorer avoids the problems although Firefox has no protected mode – but then neither did IE before Vista so that's not really a step back.

It's hard to believe there's not more on the Web when I Google "IE problems and Vista" and like terms. When I talk to other Vista users – there are some – they are also having problems. Maybe people just don't use IE when they use Vista but that seems unlikely. Maybe even fewer people use Vista Business since businesses have been reluctant to take on Vista support; so perhaps I'm more of an early adopter than is comfortable.

In response to my last rant about this, several kind readers sent suggestions; these haven't made the problems go away though.

There's supposed to be Vista service pack arriving automagically any day now. Maybe that'll help. In the meantime, I'd suggest avoiding "upgrading" to Vista.

Geothermal Installed!

 

Heating with our new electric geothermal "furnace" will cost less than half what it costs to heat with oil give current oil and electricity costs, even though electricity at $.20/Kwh here is much more expensive than in many parts of the country. This is an environmental investment which can be cost justified even without any subsidy; the whole installation was less than $20,000. Pictured below is the "furnace" – actually it's a heat pump. It's about three feet high and it's relatively noisy when it runs.

 

Here's how it works: the compressor part of the heat pump does the same thing that the compressor in an air conditioner or a refrigerator does; it squeezes a gas which cause that gas to heat up. In a refrigerator the heated gas runs through coils in the back or under the unit so it can dump that heat into the air; in that case, when the gas expands, it's cooler than it was before the whole process started and it makes the inside of your refrigerator cool but it also makes the air around the coil hot.

Since this heat pump is used for heating, the heat from compression is used in a heat exchanger to heat water. That water ends up heating the house (more on that below). Just as in the refrigerator, the gas is now cooler than it was before the whole process began because we harvested some of the energy in it to heat the water that heats the house. Water in a closed loop that runs from the heat pump over a hundred feet down into our well is used to warm the gas back up again. This exchange cools the water but, as it circulates through the well, which is a constant temperature of about 58 degrees Fahrenheit, the water warms up so that it can, in turn, warm the gas again. The well has to be deep enough or have enough flow through it or some combination of depth and flow so that it doesn't get chilled.

In warmer climates the same heat pump would be used to chill water or air and pump the excess heat back into the well. We don't have air conditioning so don't need that.

The heat pump is quite literally pumping heat out of or into the ground – that's why it's called geothermal. It does require some ground to go into; doesn't have to be a well but, if you have one, that makes it cheaper to install.

Our house is heated by circulating hot water in the floors. This heat pump can heat water to 120 degrees; but, on the coldest days, we might have to have water at 140 degrees leaving the utility room in order to assure that the furthest rooms get warmed. We could have put in an even bigger heat pump but this would have been overkill most of the time. Instead we have a small gas furnace which, when necessary (lots of computing here), steps the water up from 120 to 140. We'll experiment this winter and see how often – if at all –it's really necessary to post-heat the water. Another advantage to having the furnace is that it can run on our standby generator and keep the house warm even during a power failure. The compressor would require a monster generator because of the surge when it switches on.

The tank below is the buffer; the water flowing into it has just returned from warming the house so it, itself, has cooled. The water in this tank circulates through the heat pump which does its best to keep that water at 120 degrees. This water next goes through the furnace; so long as the water is as warm or warmer than the computer tells the furnace its output needs to be, the furnace doesn't fire.

In 2005 the United States used over 65 billion gallons of oil to heat residences. Switching to traditional radiant electrical heat is still more expensive than oil in most parts of the country – although that may well change. But geothermal electric heat use less than a third of the electricity required for radiant electrical heat. Where there is enough ground to put the coils, it's an alternative that's practical today. Look for it in much more new construction.

More on the economics of geothermal heat here although, at the time (last summer), I thought oil would only be $3.00/gallon this winter. This time next year, I hope to have some actual numbers to report.

Energy Policy – The Hard Work is Done

Mustering the political will to drive up the price of gasoline to a level where alternatives make economic sense would have been impossible – note the pandering attempts to knock it down a few cents for a few months. But the expansion of the world economy to include China and India in a big way has done that for us; there'll be ups and downs in the graph of gas and oil prices but, absent a world-wide depression, the trend line is up.

So now it's time to live with it. Hard as it is to take in the short term, higher prices for oil-based energy are a good thing in the long term. Too much money has been leaving our economy to pay for oil; the money's been going to all the wrong places; and, quite possibly, burning fossil fuels contributes to accelerated global warming.

Absent the recent run up in oil prices, we would have needed an elaborate systems of subsidies and penalties in order to get meaningful development of alternatives to oil any time soon. Now we're in a position where we can start unraveling those subsidies already in place – starting, I hope, with the subsidy for ethanol.

The trouble with subsidies is that they require the subsidizers to pick the winners and losers – always a hard thing to do with new technologies and something governments are notoriously bad at. Subsidies are also very hard to unwind. With the price of gas and oil up, private capital becomes available to finance any number of promising new technologies – there'll be some huge winners and lots of losers but that's the way technology and economies move forward. There'll be lots of parallel development.

If we want to maximize the private capital available, we have to avoid the urge to subsidize particular solutions. That's because private capital is rightly leery of competing with "free" capital from the government.

An important set of subsidies to unwind as soon as we contractually can is those that we've given to oil companies for drilling and refining. This may seem counter-productive since more domestic oil would reduce the economic and strategic harm of importing. But, if these incentives were ever needed, they're certainly not needed now with current prices for the end products. What would be really crazy – but is politically quite possible, even attractive – would be to leave the subsidies in place but also have an "excess profits tax" on oil companies which attempts to determine how those companies invest.

The subsidies for "mineral extraction" are woven deep into the tax code; I have some investments which take advantage of this. These subsidies can and should be undone legislatively, not countered with another set of taxes.

Now that the market has raised oil to a price that's relatively easy to compete with, the most important reason for not subsidizing one form of energy over another is that we'll get the "right" mix of alternatives more quickly if we let private investors take the losses and gains and explore all the niches and crannies. But, if we're not subsidizing, we also save tax money (or raise more by reducing the subsidies hidden in the tax code). That money should be used to reduce the pain for low-income working Americans by reducing social security taxes collected from worker's pay checks and making the social security trust fund whole with the foregone subsidies.

Sprint’s Big Deal – New Life for WiMax

The Wall Street Journal is reporting the terms of a yet unannounced deal which will finance a massive rollout of WiMax by a Sprint-Clearwire joint venture. Outside funding is to be provided by Intel, Google, Comcast, and Time Warner Cable as well as Bright House, a small cable company. Assuming the deal is for real, this is good news for US users of broadband and, indirectly, other users around the world.

WiMax can be used to provide both fixed and mobile voice and data services. Data speeds are up to eight times better than Verizon's EVDO (also currently offered by Sprint) and AT&T's HSPDA. Both AT&T and Verizon plan to rollout faster access – but not for another couple of years. WiMax is in use today; all it takes is money to roll it out. But it does take lots of money; the price of WiMax devices won't come down and WiMax won't be built into cellular handsets until there's some assurance of a broad market for it. That broad market is what this money provides.

Part of the WiMax promise is that it can be used both for residential and roaming data service. It's much faster than most DSL and better than what's available from cable companies today in most of the country. That means both your phone and your laptop may both connect in the same way at home, in the office, and on the road – if only someone has the money to build a national network. Now someone does.

The US broadband market needs more competition. Sprint has shown its willingness to compete on price with its flat rate unlimited voice and data plan. Now it'll be more than competitive in technology as well – at least for a while. Sprint is also active as a wholesaler supporting local providers of cellular service (as well as Amazon's Kindle). That means more marketing muscle than just Sprint alone selling this new service.

The companies providing the money are companies that have a stake in the outcome:

Intel has invested huge amounts in WiMax technology; for a while it looked like the technology might not ever get to market massively enough to get Intel's customers, the device makers, to incorporate it.

The cable companies have been offering fixed voice and Internet access for a while. But they haven't had a way to offer cellular service or mobile data. They will be able to resell Sprint's network. I'd look for that to happen in a big way. Cablecos may even use WiMax to reach subscribers who are expensive to run cable to.

Google has an interest in making sure that there are lots of competing pathways between their huge databases and those who search them. Monopoly providers would surely try to put toll booths on the search path and tap into Google's revenue.

Related posts:

Why WiMAX?

WiMAX vs. WiFi

Should a Cellular Carrier Be Your ISP ...

Google's Gigabit Gambit

 

 

 

 

Fun With Google Maps

If you have a list of locations, eventually you'll want to see them on a map – count on it.

An organization I work with gathered addresses and other useful information relevant to their mission with a web form and neatly downloaded all the information into a Google Docs spreadsheet. Eventually they decided they wanted to see it on a map. Turns out they can do that rather easily – no programming involved – by using one of the paid versions of Google Earth.

All spreadsheets can save their content as a .csv file (cells in a row are separated by commas). The paid versions of Google Earth have an import function for CSV files, which asks you which columns contain which address elements and then puts a dot on its map for each row. It works whether you have each element of the address in a separate column or have the address elements all run together as you'd enter them into a query for Google Maps. Non-address data from the spreadsheet can be displayed in the balloons for the entries when they're clicked on. Google Earth Plus is $20/year and can import up to 100 addresses at a time; Google Earth Pro is $400/year but will import as much as your machine has stomach for. Once the addresses have been loaded by a paid version of Google Earth, they can be saved as a KML file which even the free version of Google Earth can open.

There are a few drawbacks to this approach:

1. If the addresses are entered incorrectly initially in the form that gathered them, they don't get coded so they don't appear on the map. "Incorrect" is whatever the map provider Google is using says it is so some addresses are bound to get rejected no matter what.
2. You need to pay for one copy of Google Earth to create the KML files.
3. Whoever is going to view the map with the points on it needs to have Google Earth (free version is fine). However, there are web-based viewers for KML available. One is here; only hitch is I haven't gotten it to work and don't have time to debug right now. But, if this works, it gives you an easy way to make your location data available on the web.
4. Address-based geocoding is only available for certain countries.

If you can do it, it's better to gather your data on a map so the user has a chance to verify the location and correct any errors or negotiate a compromise with what the map provider expects as far as street names are concerned. That's what we did with Vermont Telecommunications Authority's "who has what broadband where" map. But collecting on a map takes some programming and may not be worth it for your project. With limitations, you can gather data on a form and make it visible on a map – no programming required.

It IS Pandering

The proposed gas tax holiday IS pandering – and bad economics to boot. Good for Barack Obama for saying so; bad for John McCain (whom I currently plan to vote for) for proposing it; bad for Hillary Clinton for supporting the proposal even though it helps her push Obama into the "elitist" corner.

If you assume that you get the whole 18.4 cents on every gallon you buy for two months – you won't end up with a whole lot of extra money in your pocket unless you plan to do a whole lot of driving. One thousand miles/month is well over the national average; that's 50 gallons at 20 miles per gallon; over two months you could save $18.40 – barely enough to take the family to McDonald's once not even counting the gas to get there. On the other hand, assuming gas has hit $4.00/gallon by then, if you saved 10% of the gas you would normally burn either by driving slower and/or driving less, you'd save $40 and none of that is gonna end up in any coffers but yours.

Of course, the $18.40 isn't really going to get to consumers. Retail prices don't go down as fast as they go up. To some extent, we're being charged what we're willing to pay for gas. Most economists agree that much if not all of the forgone federal tax revenue will end up sticking somewhere in the supply chain – some could even get all the way back to the countries that produce oil.

I do think it's elitist to say, as some environmentalists are, that gas prices ought to be pushed up even further to discourage consumption. I was for a gas tax increase for just that purpose when prices were a $1.50/gallon less than they are today. We've had enough discouragement for now; we are responding by driving less and buying smaller cars. As important as that response is for national security (or the environment, take your pick), it isn't going to do more than slow the increase in gas prices as Chinese and Indians drive into the middle class. The cost of driving may never come down from where it is today; it won't even stabilize until it becomes an all-electric experience for many people.

Wouldn't it be great if someone would just skip the campaign rhetoric and debates about debates and pandering and spell out a real energy policy? Between those of us suffering from higher energy prices and those of us concerned about the national security and economic implications of sending money to unstable oil-producing places and out of here and those of us concerned about global warming, it really would be possible to rally a consensus around much more than band-aids.

BTW, is it just me or has anyone else noticed that gas pumps seem to be dispensing gas more slowly even though the dollars roll up quicker? It can't be that their speed is measured in dollars per minute and has reached its limit; but it sure seems that way. Makes the pain even worse.

JavaScript or PHP – Why It Sometimes Matters

It's all about whose computing power gets used – yours the application provider or yours the end user. If you're the user, you probably don't care since you are waiting for your browser to show you something in either case and you already paid for your computer. If you're the application developer, you may care a lot because you have to pay for server processing cycles one way or the other but you don't have to pay for user machine cycles.

If you are not involved in providing web applications in any way, you may want to bail out of the rest of this post at the end of this paragraph. Before you go an interesting factoid is that PHP stands for "PHP: Hypertext Preprocessor" – the acronym is recursive. Of course, if you're not a nerd you may not think this is interesting so I'll give you one more factoid. You may have noticed that many web addresses end with ".php". These are web pages that are created on the fly on servers using PHP. Pages which are more static OR are made dynamic on the user machine often end with ".htm" or ".html". Now you know.

OK; only us nerds are left now. You all know that a page which is pure HTML is pretty static; appropriate for lots of things but pretty static. But you also know that you can add some JavaScript and have the page become interactive; even better, you can use AJAX (Asynchronous JavaScript and XML) and load all kinds of data into an existing page to create an interesting and dynamic application. This data doesn't even have to be XML.

Pretty much anything you can do with JavaScript on a user's computer you can also do with PHP on your own server. The two languages are similar so the complexity's about the same either way. One huge exception is that you can't do anything with PHP that requires lightening fast interaction like reacting to a mouse-over, a click, or a key-press instantaneously. All PHP interactions gotta go from the user's computer to the host and back and that takes noticeable time.

One thing that you can do in PHP on your server that you can't do in JavaScript is load data or XML from domains other than the one the current page or iFrame was served from. That can be a significant limitation. You also can't keep your code secret if it's JavaScript; it's there in source form on every user's computer. You may want to keep it secret either because it's where your value comes from OR because it would be a security risk to you or the user to expose the code.

Many APIs like the Facebook API or the Amazon S3 API assume that you have a server at your disposal. I did manage to use S3 and maintain necessary security with only a client application and no server but had to jump through hoops to do it (see Amazon S3 – Backstory for Nerds - Part 1 and Amazon S3 – Backstory for Nerds – Part 2). 

In many ways it's easier to test and debug using PHP than JavaScript. The problem with JavaScript is that different browsers treat it differently. You sometimes have to ask in the code what browser you're supporting. You have to do extensive cross-browser testing. I spent (wasted) most of today on a problem caused by the fact that Firefox fills in the "Content-Type" header in an HTTPRequest and IE doesn't. The PHP you run on your server is going to execute the same way no matter what browser the user has – of course, the resultant HTML may not image the same way but that's a problem no matter how you do you development.

You can mix JavaScript and PHP so you can use each to get around the limitations of the other.

That gets us back to whose CPU cycles do you want to burn. If you're planning to have a blockbuster virally-spreading world-changing application (which, of course, you aren't going to charge for) then you don't want to have to buy a huge server farm. You can outsource to Amazon EC2 or perhaps use the Google apps server; but millions of users are still gonna be expensive to support. That'll often be a good reason to use JavaScript and the user's cycles rather than PHP and cycles you pay for. The user isn't hurt by this because she's already waiting for her machine to do something and just wants it done fast – it may, in fact, happen faster with local JavaScript execution than if an interaction with the server is required.

The P2P services, including Skype and BitTorrent and Napster, all rely on user resources – both bandwidth and computing power. My guess is that many web services running in browsers will be successful because scalability is instant and "free" when each user's machine provide the resources to handle most of the incremental load that user represents and when the bandwidth for communication between application and server is used sparingly. That's a big reason for doing as much in JavaScript as you can.

Vista Misery and Mysteries

Wasn't planning to install Vista on my new Toughbook CF-30 because it's been hard to deal with on Mary's new HP; but I did and now have both some misery and an Internet Explorer protected mode mystery which I'd be glad for nerd help with.

I liked the fact that the new Toughbook came with XP installed plus recover DVDs for both Vista Business and XP. That meant, I thought, that I could give Vista a while to stabilize and then install it. Unlike Mary's preinstalled Vista Home, I knew I could even uninstall Vista and reinstall XP since I have the business edition and Panasonic supplied me with both recover DVDs including all the crucial drivers needed to go back and forth.

But then the fine print: you have to wipe your hard drive of all content to install Vista. Do I want to spend weeks getting things the way I want them on a new machine, run a couple of months with XP, then start all over again with Vista and a "clean" machine? No, I decided, since I have nothing of mine on the machine now, this is the time to install Vista. Then I'll get the Vista version of everything.

Right now I'm regretting the decision. And looking for nerd help with a mystery.

Vista is running and isn't noticeably slow on my new machine; it hasn't crashed. That's the good news.

But Vista seems determined to protect me from myself even though I run with administrator privileges.

At first I couldn't get any ActiveX extensions to install. The yellow bar above the browser window which usually warns me that I've clicked on something which wants to install an ActiveX extension now didn't give me the options of installing; it just told me that my security settings didn't permit ActiveX extensions to be installed; this despite the fact that I'd deliberately clicked an option to prompt for permission before installing an extension. Couldn't even get Microsoft's own software verification extension to install to get the latest fixes to Office 2007.

Turning off "Protected Mode" in the browser let extensions install (but without a warning which I don't like either). Then, while trying to figure out why there were no time-wasting games like minesweeper around, I discovered (by Googling, of course) that you can go through Control Panel/Programs/Programs and Features to "Turn Windows feature on and off". This not only lets you turn games on; it also lets you turn on the ActiveX installer service. I've verified that games instantly appeared; haven't stumbled across an uninstalled ActiveX component since so not sure this is working properly.

But here's the mystery:

Web pages that used to work fine including basic Facebook pages now SOMETIMES break because, according to IE, it can't load a DLL (doesn't say which DLL). This never happens when Protected Mode is turned off. It doesn't get cured by a reload but sometimes the same page WILL load without an error much later even when I know the HTML of the page hasn't changed. It never happens in Firefox but Firefox doesn't have a Protected Mode. If it happened in Firefox, it'd be easier to debug because of Firebug. It doesn't seem to happen in Protected Mode on Mary's machine because some of these are pages she goes to often and she hasn't complained (but she's running Vista Home). I can't run a parallel test or compare all the settings on our machines because she and her machine are traveling.

It's not a solution for me to just run Firefox or run in unprotected mode because, as a developer, I need to know why pages sometimes break. If they break for me, they'll break for other people as well.

Haven't been able to Google my way to a solution or find one on Microsoft's site.

Ideas anyone?

Why Old Computers Get Slow

My Toughbook CF-29, bought back in 2004 when I left the corporate world and expected to spend many days in physically tough places, has been getting slow. Mouse clicks and resulting actions are increasingly far apart. Windows open blank and don't fill in for eons. Programs go into non-responding status; sometimes they recover; sometimes they don't. Rebooting temporarily speeds things up but it takes longer and longer to shut down and restart.

Why? You ask. The speed of light hasn't changed. The circuits can't be slower than on the day you bought the thing. Why are old computers slow computers.

A small part of the problem is barnacles. Most of the problem is that programs are written for the average two year old machine and its capabilities. The old computer bogs down under limits of memory and processing power when asked to do tasks designed for its more modern successors.

The barnacles are all the stuff you installed over the years and probably aren't even using anymore. Inactive programs don't do any harm except take up disk space but some of what you installed, printer and communication programs and drivers for example, have a small part which loads itself at startup and absorbs resources all the time the machine is running. With some trouble (and some tuneup utilities I've never tried), you can find these and turn them off. If you don't, they absorb more and more memory and processing power. There is some "printer subsystem" on my old Toughbook which reports its own failure every couple of hours and asks permission to tell Microsoft about the problem. Microsoft doesn't offer any solutions when told and the failure of this subsystem doesn't seem to affect my printing; but it's there somewhere.

When I bought the old Toughbook I installed as much RAM (memory) as it would take: 512 megabytes; seemed huge to a guy who wrote programs for the 128 kilobyte Mac (the first Mac) and 16kb TRS-80 model 1. (a megabyte is a thousand kilobytes). "My" first computer was an IBM 7090 the size of a basketball court with the equivalent of about 156kB. It cost millions of dollars. But I've digressed to my age… back to the old computer.

The CF-29 has a single processor which runs (walks by today's standards) at 1.3GHz. It has a 40 gigabyte hard drive which is almost full.

When I first used the machine I had one window open on the Internet most of the time. Now I have at least six tabs open in Explorer – because I can. I've gotten a bigger screen and, when at my desk, I have windows spread across two screens for easier cross reference and cutting and pasting or perhaps because I have code in one window and I'm watching how it runs in another window. Whatever, I've got more stuff running at one time.

There isn't enough memory on my machine for all this stuff to be running at once so Windows puts some of it in "virtual memory" – really on disk from which it must be reloaded before use. Often it is clear that my computer is "thrashing" – the technical term for swapping pages in and out of memory so often that it can't do any real work. Picture a hundred people trying to work on something in a room with only space for ten; every time someone else is needed for the task at hand; someone has push his way out so someone else can push her way in. But the guy who went out hadn't finished so he reenters and pushes someone else out. Pretty soon it's all elbows and no productivity. That's my computer.

Programs like Google Earth and Sketchup assume that my computer can draw lightening fast. It can't at the resolutions they're feeding it. Turning the scroll wheel for zooming a graphic image results in a jerky ascent or descent like a rocket with misfiring engines.

Web pages now contain incredibly elaborate Javascript – easy to write; cool to use; but a big burden on an old CPU to interpret and execute. Then there's Flash. And videos embedded in everything. You can almost hear that lonely old CPU groan.

So that's why it's time for a new computer. My new Toughbook CF-30 (I still plan to go to all those tough places) has 4 gigabytes (4 billion bytes) of RAM (although Windows can only see 3.3 gig). It has an 80 gig hard drive. It has TWO processors each screaming along at 1.6GHz. It has more auxiliary graphics processors to take the load off the main processor. It has an embedded EV-DO radio and GPS to reduce the clutter in my gadget bag.

And it has Vista… So far not at all to my liking.

Testing Office 2007

Meant to start this series on moving to a new computer (gulp) and upgrading to Office 2007 (gulp, gulp) and Vista Pro (GULP!) in an orderly way with a post on how computers get old and slow. But when I told the new Microsoft Word that comes with Office 2007 that I wanted to create a new document, it asked me if I really meant a document or a blog post. Well, this is meant to be a blog post so…

Word asked me, reasonably, who my blog provider is – it's TypePad – and then for my TypePad ID and password; it warned me that these would be transmitted in the clear. Since I have several test blogs on TypePad as well as Fractals of Change, Word asked which one I wanted. Maybe I should have started with one of the test blogs but what the hell… let's give it a try.

If you read this post, it is possible to go directly from Word to a post without cutting and pasting. I'll let you know what steps intervene after I click the Publish icon.

Reset#1: After I clicked Publish, I noticed two typos and that there is an Insert Category button. Clicked cancel. Word became unresponsive for a while according to Vista but did recover and say that the blog service provider wasn't responding.

Clicked Insert Category; got back my category list from Typepad (nice) but can only choose one category even though TypePad makes it possible to choose several (not nice) but will continue the experiment.

Reset#2: OK. It did publish; in fact, so did my previous attempt. There is an Open Existing option so have done that and now am presumably editing and able to repost. Tried to use this option to delete the earlier post but can't do that (no big deal). Will try reposting by clicking Publish again. Then will go directly into TypePad to add more categories and Technorati Tags (too bad but not critical).

Just saw an option for handling pictures (imbedded, I hope) by uploading them to the blog provider (default) or somewhere else (which the documentation says can be anything with a public URL including Flickr). Will insert a picture below and leave the default to have it hosted at TypePad. Hmm.. scaling only gave a choice of 1% (maybe for a thumbnail?) but could simply specify a new height and the width adjusted to keep the aspect ratio. We'll see what happens.

 

Vermont Files in Support of Using White Space for Mobile Broadband Access

The Vermont Public Service Department and the Vermont Telecommunications Authority have joined in an ex parte filing at the Federal Communications Commission urging that the Commission “move expeditiously to adopt the necessary technical parameters … and help make this promising technology [use of the so-called ‘TV whitespaces’] a reality.” Given that the docket has been open since May of 2004, a little expeditiousness is certainly in order.

“TV white spaces” is the term used by the FCC but it’s a misnomer; no broadcaster has actually paid for any of the spectrum at issue; no one is using it; in short; it’s wasted. Originally, before cable and satellite TV and before the Internet, it was reasonably believed that this spectrum would eventually be occupied by a proliferation of over-the-air stations. That’s not gonna happen. Vermont has as much radio spectrum “reserved” for over-the-air TV stations as New York City – 50 channels worth. That “reserved” spectrum is not of any use to anyone and won’t be until the FCC promulgates some rules for its use.

The filing explains the many reasons why this spectrum is ideally suited to meeting the needs or rural America for much better broadband and cellular coverage:

“First, rural areas like Vermont have relatively fewer TV broadcasters and therefore more unused ‘white spaces.’ Moreover, rural communities also have the largest geographic areas without access to wireless services. Second, the ability of TV frequencies to propagate over great distances and difficult terrain provides an opportunity to reach locations too economically challenging for existing wireless services. Third, the use of TV ‘white space’ for the provision of rural broadband is an alternative means of accomplishing the Commission’s universal service goal of deploying advanced services to all areas of the nation without requiring additional funding mechanisms. In fact, the use of TV ‘white space’ could actually decrease the demand for universal service funding at a time when the level of funding is facing heightened scrutiny.”

The filing makes clear that the petitioners do NOT think that this spectrum should be auctioned off at a high price. The greatest public good will come from making these public resources available “at low or no-cost to those entities willing to utilize them for such purpose [broadband and mobile access].”

It will take the concentrated political power of rural America to free up this spectrum to meet the rural need for better communication. But this isn’t urban vs. rural; urban areas also have something to gain from better spectrum availability and nothing to lose.

Not to over-dramatize but I see this as the public interest vs. entrenched communications interests. The TV industry would like to sit on this spectrum without paying for it “just in case”; they also may be worried about Internet use of the spectrum becoming a competing “channel” for delivering entertainment. Traditional communications carriers benefit from LACK of competition in the US broadband market; they have no reason to want to see competition growing like weeds (or, more accurately, like WiFi) in fields of open spectrum.

Google and other “Internet” companies do have an interest in keeping their paths to the consumer unblocked; competition would be good for that. This post is about a proposal Google has made for putting the unused white space to work.

Disclosure: My wife, Mary Evslin, is Chair of the Vermont Telecommunications Authority.

Auditioning for New Roles

Below I am practicing to supplant pigs and toasters as a screensaver.

In case that doesn't work, I could always be a demo dolly (what's the masculine of that?). I'll buy a BUG like the one I'm demoing below for anyone who can correctly identify the polo shirt I'm wearing in this picture before the end of April.

If all else fails, it's back to email.

 

All pictures are from Kinnernet 2008. Great fun as usual.

Cordless or Wireless? Good Question

We don’t do spring here in Vermont so now it’s summer although there’s still plenty of snow gleaming in the mountains. Time for me to work outside (at least until we get our obligatory late season snowstorm).

My cordless phone is scratchy by the time I get out on the deck; it doesn’t like being that far from the base station. The WiFi isn’t great on the deck either.

I could move the base station for the cordless phone and install a repeater for the WiFi signal. Last year I used my antenna and high-power WiFi card to make WiFi work right outside. But that’s all a lot of trouble.

Instead I made my calls using my wireless (aka cellular) phone. And I put the EVDO USB modem in my computer and just used that for connectivity. Since I never use my 500 wireless minutes each month nor the 5 gigabytes per month included in my EVDO account, it doesn’t cost me anything incremental to be in my travel configuration while on the deck; and it’s a lot more convenient than making the house radios have good coverage outside.

So here’s the question: will there come a time when we don’t install our own little radios for voice and data at the end of the wires, cables, or fiber that comes into our houses? Will we just pick up the same signal from our carriers that we use when we’re traveling inside the house as well as on the deck and in the car?

Clearly WON’T happen unless the carriers lower the prices for cellular and EVDO and lift the volume limits. $99/month for unlimited talking on Verizon Wireless or AT&T is a lot more than $24.95 on Vonage which also includes reasonable rates on international calling. 5 gig would disappear pretty soon if I were doing my nightly over-the-net backups and watching MLB.com on EVDO. Moreover EVDO isn’t really fast enough for lots of web stuff.

The conventional wisdom is that eventually voice and data will come over a fiber into the house and then be distributed wirelessly thoughout the house and maybe the yard and that mobile needs will continue to be met by different technology at a higher price. Maybe the conventional wisdom is right but it’s always worth questioning.

I think there’s a strong probability that not just the last 100 feet but the whole last mile will be wireless in many places. Radio technology is advancing very quickly. There would be plenty of spectrum IFF (and it’s a big IFF) there were regulatory reform to allow use of whitespace and make much more spectrum open. As we (and our computers) spend more and more time connected, we’ll be more and more impatient with having to switch connectivity modes when we walk out the front door.

That would mean no communication wires, cables or fibers coming to most single family residences. That could also mean true competition in communication services just as cell phone service offers more choices, more competition, and more innovation than landline service does today. It’s hard to make a business case for duplicate networks to each house; much easier to make the case for competitive radios, even on the same towers.

Just a speculation.

What Would Success Look Like in Iraq?

How about an end to private militias? That’s pretty important to the kind of success in Iraq that lets most American troops leave knowing that they’ve helped make the world a safer place. How about a Shiite-led government cracking down on Shiite militias while Sunni tribes turn on Sunni al Qaeda? That would be pretty good.

And how about an Iraqi army that quickly corrects its mistakes, flushes out its non-performers, tries again and succeeds? An Iraqi army that can provide security for Iraqis by itself? Wow, that would be great.

There’s plenty of reason to be cynical about apparent good news coming out of Iraq. All of the many sides (including ours) have good reasons and bad to spin the facts as much as they can, especially in the run-up to the American presidential election. Nevertheless, there are reasons to be hopeful as well, whether or not you think we should have invaded in the first place.

If it was a bad sign that Iraqi government forces didn’t flush the Mehdi army from Basra when they tried a few weeks ago, it’s got to be a good sign that they regrouped and apparently succeeded rather easily at taking control of the city this time. Sure, they had help from the US and UK; but there isn’t any question that the ground forces were overwhelmingly Iraqi.

The fact that Muqtada al-Sadr, apparently somewhere in Iran, claims to have ordered his forces to surrender their Basra headquarters indicates that that he didn’t have much of an option. His threats to end the truce he formerly “proclaimed” while his people are already fighting in Sadr City, Nasiriya, and, until very recently, Basra mean either that he has little control over his forces anyway or that he doesn’t have the means for a stronger counter-attack (I could be tragically wrong about this, of course; but I hope not). It’s a good thing that this very anti-American cleric seems to be losing power.

It’s puzzling that Iran praised the Iraqi government’s action against al-Sadr, whom they shelter and probably helped arm. That could be an ominous sign that the al-Maliki government may be too much under the control of Iran. Or it could simply mean that, for the moment, US and Iranian interests happen to coincide – probably not something we can build on.

All the many sides have their own reasons for continuing the power struggle in Iraq – not least among them that none of them can be sure of their fate if the other guys win. At a strategic level – which probably doesn’t matter much on the dangerous streets, the importance of influencing the American electorate is huge. If the situation appears to be deteriorating further, it’s likely that the next president will be someone committed to a rapid pullback of American troops; if the aftermath of the American surge is what looks like real progress (which eventually means an orderly withdrawal of most troops), then the next president may be McCain who will keep up the fight. (I’m not making a political statement; just analyzing the situation. I actually think we have a lot of important issues besides Iraq which ought to determine who is our next president.)

It’s important that we be as shrewd as we can be in recognizing both failure and success in Iraq. The original invasion was a huge success; the aftermath was a dismal failure of both planning and execution; the surge appears to have been helpful; now what’s happening? Our press needs to be as objective as human beings can be. We can anticipate efforts by those hostile to our presence to influence our election with a violent “surge” of their own which will be, like the Tet Offensive so long ago, either a sign of enemy desperation or of our inability to control events on the ground.

I wouldn’t presume to say how this story ends but am watching with an open mind and a little bit of hope.

FeedBlitz, Gawker Media, and Amazon S3

FeedBlitz (in which I’m an investor and board member) recently announced that all Gawker Media sites are offering weekly updates of their top five stories via FeedBlitz-generated email. Gawker needed to be able to customize the look-and-feel of the newsletters which go out to match the appearance of the individual sites and needed to be able to let readers manage their own signup, email address changes, and unsignup; FeedBlitz is an answer to these needs.

Not coincidentally, FeedBlitz CEO Phil Hollows also blogged that FeedBlitz has begun using Amazon S3 for many of its storage needs. Huge potential traffic from Gawker properties and similar megasites makes it essential to be able to scale fast but impossible to predict how fast. S3 is an answer to these needs (I have no financial interest in S3 but am fascinated by its potential to enable rapid prototyping and scaling).

Phil describes Gawker this way: “Gawker sites… are well written but often irreverent, somewhat profane, sometimes politically incorrect and frequently deal with topics you might not want to discuss with your mother.” They include Valleywag, Lifehacker, Gizmodo, and Defamer among many others and are, to say the least, often visited. To see how closely Gawker was able to reproduce the graphic elements of the websites in the automatically generated email, look here for a Valleywag sample and here for Gizmodo.

FeedBlitz is sending out more than 3 million emails (double opt-in emails, not spam) per day. In the old days when a company grew, it brought operations inhouse in order to save money – it verticalized. That was then and this is now. The fastest way for FeedBlitz or any other modern company to grow is to outsource everything that’s not a core competence. So FeedBlitz moved all of its image and script-serving to Amazon S3 rather than just keep buying bigger and bigger servers. Total bill at this point: about $3/day for a significant amount of both storage and traffic. And Amazon’s multiple connections to the Internet backbone and replicated database can serve this stuff up much faster than any local hosting site or a (shudder) in-company data center. No matter how fast FeedBlitz grows, its growth will be easily absorbed within the Amazon cloud. FeedBlitz engineering is free to concentrate on adding new functionality for publishers of newsletters as it worries less about pure scaling of existing functionality – often an Achilles heel for fast-growing services.

Phil says: “If you're running a site or service that is going to get big, I'm now of the opinion that you're nuts not to outsource to S3 or a similar service to store and serve objects that aren't core to your value add. It's faster, better and cheaper and whole lot less hassle. Do it!”

Third Life: Social Networking Breakfast with Jeff Pulver

“It’s like Facebook only it’s not online,” someone at Jeff Pulver’s social networking breakfast in Tel Aviv explained to somebody else. “There’s tagging and everything but it’s not virtual.” That’s a pretty accurate description of these real world events which build on not only connections but also techniques learned online.