Internet Plumbing - Some thoughts

I was reading a Scientific American article on how the United States has fallen behind on Internet adoption and speeds:

Why Broadband Service in the U.S. Is So Awful

There's a similar, older article:

http://www.washingtonpost.com/wp-dyn/content/article/2010/04/16/AR2010041601316.html

but along the same theme, that we need high-speed internet or broadband in the U.S., and in many instances is unavailable or too slow or too expensive, and the FCC and our regulatory system are to blame.  I guess I agree in part, but I see it as more nuanced.

I have broadband, it's through Time Warner Cable, mostly it "works", but it can get kind of slow (and yesterday at least, even the HD TV channel I was trying to watch was having problems), so even the 1MbPS I might normally see can grind down to something a fraction of that.  It is a characteristic of the implementation, cable channel bandwidth used to carry the IP traffic for a neighborhood kind of gets saturated, things can kind of slow down. I should note in fairness to Time Warner, at times the speeds are very good, so it's kind of a supply and demand thing. Generally Netlflix (IP Video on demand) viewing can be problematical, and often the resolution, particularly for anything with a lot of action, can be less than optimal, especially in HD (Blu-ray has nothing to worry about, at least for now).  DSL is offered, but doesn't seem to work in our neighborhood, too far from the central station or something marginal in the local phone network, next-door neighbor basically gave up.  And Verizon FIOS ends about 500 meters from my house, where the lines would need to go underground, and I've been waiting for years to have them as an alternative, but they just have not gotten around to it, and it appears they're winding down their new deployments, at least for now.  So for better, or worse, I'm stuck with Time Warner.  My experience, in general, is about the same set of problems as described in the article, and other places, of a typical U.S. based cable internet customer.

This all got me to thinking (yes, dangerous doing that sometimes).  I'm not sure I'm going to blame the FCC, oh they don't get totally off the hook either as they haven't been doing a very good job, but the problem is not really caused by them, and we are all feeling our way on how best to adopt to the new technologies.  To me, this is a problem where we have not in the U.S. found a model or a set of regulations that encourage the fast and cheap delivery of internet.  The technology is pretty clearly there, I run a 1GbPS internal house network in my house (1000x faster than the nominal speed coming in), it's not Asymmetric at all, and the 16 port switch it's on cost lest than $100 (that's less than 2 months of my internet bill).  High speed, cheap, ethernet is available for maybe $25 a port, total.  I'm not running a managed switch, but it's not bottom of the line, either.  But actually, the incentives for the cable company are to not provide cheap and unlimited high speed internet, when you think about it.  First, if you're a monopoly, you charge what the market will bear, not a penny less.  Second, if there was uncapped, net neutral type internet connection were available, customers would switch away from TWC's own offerings for video on demand, etc..  So how do we  fix this?  What do we look for by the FCC or at least the U.S. government?  We have to come up with the right regulations or laws to put the incentives back into providing the fast internet, something along these lines:

1. Don't allow Internet Caps (maxes per month, so long as the usage is legitimate or protected)
2. Require Net-Neutrality (or specify that it's made available for a nominal extra charge)
3. Standardized offerings (let the industry work out and specify the grades, but something like:)
• Basic Broadband:  Minimum of 1MbPS download, 100kbPS upload, 98% of the time
• Enhanced Broadband:  Minimum of 2MbPS down/200kbPS up
• Fast Broadband:  10MbPS up and down (Fast Ethernet)
• Very Fast Broadband:  1GbPS up and down (Gb Ethernet)
• Business Class Broadband:  10GbPS up and down (Ethernet)
4. The company could make non-standard offerings, but it must offer at least a subset of the above standard offerings, and it must either provide either a standard ethernet connection, or a box to convert to that, as part of its pricing.  All advertisements that show or discuss the non-standard offerings must show the standard ones and pricing in a legible (10 pt or higher) table, or a URL directly to the same.
5. Don't allow localities to add charges or prevent internet connections via taxes, local laws, or rules, including right of way, at least for Standardized offerings (interstate commerce based laws, since the internet is used frequently for interstate, national, international communications and commerce) 
6. In areas where there was no competition in Broadband, from at least 3 providers, prices and offerings would have regulated maximums, set by the FCC, based on housing density.  
7. As an incentive, the U.S. government would subsidize the installation of high speed internet to Schools and rural or semi-rural areas, or other connections that would not be financially feasible otherwise, companies or individuals could apply for a grant to connect to internet hubs, but any subsidized connections would be open after a recovery period to other providers. 

It's probably the last three that are the key to real competition.  We have to come up with a system of incentives and a set of rules eliminating barriers to higher speed internet.  If people can pay more and get more, they'll do that.  It's pretty clear that the internet connections will become over time (and for competitiveness, perhaps already have) more important to consumers than telephones (which were highly regulated) and TVs (moderately regulated), as the internet can provide both.  We need to find a way in the U.S. that anyone can get an internet connection, at steadily higher infrastructure speeds and feeds, reasonable latencies.  To that extent, it becomes a little like the Finnish model:

CNN: Fast Internet access becomes a legal right in Finland

This is really where we, in the U.S., also need to end up.  I think we in the U.S. are sometimes too complacent, assuming that a U.S. invention will be something we're always ahead on.  The world doesn't work that way, you have to keep working and innovating to stay ahead.  The trick is to find the right set of incentives so that our capitalist system manages to get there quickly and efficiently.

eBooks - Some thoughts

First, fair warning, I own an Amazon Kindle eBook Reader, and while I dislike DRM, I don't see Amazon as the root-of-all-evil for using it (anymore than I saw Apple as the same for music and the iPod).  As a consumer, I'll buy whatever seems to work the best, and not worry too much if the there's not perfection there -- because perfection I find to be elusive and at best transient.  And also fair warning, I read a LOT of books, I have shelves and shelves at home (and work) groaning from stacks of books, I spend a lot on books, I love to read books, multiple areas, my Mom worked in a library, so I am not your average consumer in that area, I'm probably 99+% percentile.  If they ever get RFID working for humans, flashing blue lights will go off whenever I stumble into a Barnes & Noble or Softpro or Half-Price Books or other bookstore.  So all of that out of the way, let me put forward some thoughts on eBooks.

eBooks are Great!

1. They're portable!
2. You can carry a lot of books in essentially the eReader space of one printed book
3. You can read a book (or at least an Amazon Kindle book) on all sorts of platforms (PCs, Web, eReaders, Smartphones, etc.)
4. eBooks are easier to buy (no having to find time to run over to B&N or Quantum or wherever, chasing down or ordering a book, you see something you like in the NYT books section or wherever, you put it on your wish-list, or just buy it and done)
5. eReaders prices have come down to reasonable levels -- about what it costs me for a good bookshelf for maybe 1/10th the number of books the reader can hold
6. eBooks are about as Green as you can get, save a tree! Virtually carbon-neutral.

eBooks are not-so Great!

1. Screens are too small (6 or 9.7" typically -- ok for fiction, not so good for technical or anything with a lot of pictures or graphics)
2. Color needs work (someone's going to say iPad, but I see that as just a start, and you still need to solve #1, so nobody is going to say it's to the point that it does an art-book or coffee-table type book very well, close but not quite)
3. DRM and Closed Platforms (OK, I did say I could live with DRM, but I still don't like it, and the Kindle/ePub wars are sort of Sony/VHS, getting tired of these things, there should be a standard)
4. Not all books are available

which leads me to really one of the biggest impediments, right now, ePublishers:


Publishers don't seem to Get eBooks

There are exceptions of course, O'Reilly seems to get eBooks, but on the whole, despite the fact that most of the Web is words, and publishers have all along been about making money by making words available to the public, many publishers seem to be caught back in some time warp where many would just like to pretend eBooks don't exist.  This is not unprecedented, look at how much trouble Apple had with music publishers and acceptance of the iPod model, and that was even after digital copying (and some dumb pricing models) had pretty much started to destroy their market; Apple was offering them a lifeline and they were slow to grab hold of it.  Amazon (and to be fair, others) were giving book publishers the same sort of lifeline, creating a digital book distribution system for published works, in effect eliminating 90% of the distribution costs, eliminating returns and remainders, improving being able to keep hot-sellers in stock, virtually eliminating the half-price seller and library book sale competition, etc.  But the publishers got too caught up in losing pricing control, and forced Amazon over into a controlled pricing model by using Apple and selective (almost monopolistic) withdrawal threats.  As a consumer, parts of what they did I took a very dim view of.  So I think there should be a ePublisher Bill-of-Rights, along these lines:

1. All books will be published in eBook formats for all the major platforms
2. eBooks will always be priced at less than the discounted going paper book price,
   that's about 30-40% off for a hardcover, about 10% off of a large paperback, never more than the mass market paperback if that comes out -- all to reflect the lower distribution costs of eBooks

I realize change is traumatizing to the Publishers, but change is not apriori bad, it can re-energize an industry.  The main problem I think with the music (and to some degree movie) industry and digital transformation is that they were slow to respond and then did so in ways where they tried to restrict the market or got maybe a little too greedy.  You're borderline greedy with the forced higher prices now, and as a consumer, I really don't like it when you come out with a paperback, and don't adjust the eBook prices to match.  Think of what happened to the music industry when CDs came along, they made a lot of money selling or reselling in the different format, book Publishers could too.  Customers will applaud you for it, if you handle it correctly.

Xmarks and Web Application Funding

The nature of what I do (for work and home) involves some Web development, using multiple browsers (FireFox, Chrome, Safari, Internet Explorer, Opera, etc.) on multiple platforms, workstations and PCs, and the best tool I've found so far to manage all the Bookmarks links is Xmarks.  

http://www.xmarks.com

While the Google toolbar isn't bad, and the built in synchronizations in Firefox and Chrome are OK, I've found that the platforms and synchronization capabilities in Xmarks to be a lot better.  The cross browser synchronization is where it particularly stands out from the pack.  Update in one browser on one PC, and poof, the bookmarks are updated in other browsers on the same or other PCs (or Macs), all without giving up any toolbar (screen space) to speak of.

And it is Free!  Or at least has been.  You cannot really beat free as a consumer.  Oh, I'd pay for it, took the pledge actually, but that brings me to the funding model problem on Web Applications.  In a traditional Application environment, you buy the computer, you buy the software to get it running, you buy Applications on top to do what you need to do.  The hardware and so forth frequently will need updating, so there's an ongoing maintenance cost, but mostly it stays running.  Web Applications offer the advantage of doing a lot of that for you.  Somebody else worries the details, costs are optimized and spread out against a bigger community.  I wouldn't use Web applications for something really secure or high performance (though the Cloud stuff by IBM, Google, Amazon, etc. seems off trying to solve some of that) or where I was basically paying for a server or server farm, but for small, non-critcal types of things, the Web Application (and eventually Cloud) model should work very well.  It's clearly the future.

But then, and this is the nasty bit, what to do when a Web application you use and love potentially goes away.  They have to reach a critical mass, they have to find an ongoing business or funding model.  And that's where Xmarks is apparently struggling.  It's a great, really industry-best application, gives you some (browser) vendor independence, oil-in-the-engine type of thing for the Web, but it is tough to figure out how to pay for it and make some money.  So a start-up on Xmarks must make a transition, become self-sustaining, or get bought by another outfit which makes them part of a bigger whole.  Hopefully, someone with deeper pockets will buy Xmarks, the most likely candidates would probably be ironically be Google or Yahoo (the crosss browser capabilities being of most interest there, despite Chorme (which is a browser I love, but not very big yet in marketshare)), or even Mozilla (but they don't have deep pockets), or perhaps IBM or Oracle (which do).  There's an actual huge intrinsic value in the data in Bookmark farms.  Hopefully that will happen, it should, there's a bigger picture aspect to all of this.  However, for Web application users in general?  That's the tough one.

The Basics....

I need some good example of Boilerplate to do this properly, but let's start with:

• The opinions expressed in this Blog are my own, and should in no way be interpreted as representing the opinions of others, or of the company I work for
• I will try and keep observations to facts as I know or understand them, but there may be errors from time to time, and while I apologize should this occur, I assume no liability express or implied, and remedial actions will be limited to posting corrections or withdrawing the post.

Thanks, and Enjoy!