About Tom Evslin

Video Profile of Tom Evslin

Follow Tom Evslin on Twitter


Add to Technorati Favorites!
Powered by TypePad
Member since 01/2005


Hey NYTimes, Where’s the Link?

It's no longer acceptable online to make a naked assertion on a controversial subject without at least a link to back it up, at least not for anyone practicing serious journalism and writing a news story. But here's what appeared in the New York Times online today in a story by John Broder headlined Energy Dept. Panel to Revise Standards for Gas Extraction:

"…the practice [hydraulic fracturing] also pours millions of gallons of dangerous chemicals into the ground and into wastewater treatment systems, which in some cases cannot remove all the potential toxins. There are also numerous documented cases in which fracking fluids leaked into aquifers [emphasis mine] and contaminated drinking water."

The article doesn't cite any of the "numerous documented cases"; it doesn't link to any source for this assertion. Although there have been allegations that fracking fluids leak into aquifers, there are good reasons why this is unlikely (see here) and I've never seen a "documented" case that it happened. I also believe he actually means millions of gallons of water which contain some toxic chemicals. Now I can be wrong and would like to know it if I am. I would've liked to follow a link from the NYTimes story to some documentation. But there is no link.

I understand why it's hard to put citations in print stories and this story also did appear in the paper edition. But paper now needs to be treated as the degenerate form of the story – the web makes it easy to link and easy for a reader to check sources and get more information. Links belong in online stories.

I modify my posts for print and sometimes have to remove links. If I think they're important for credibility, I put the text of the link into the article, although sometimes editors take those long strings out. If we're going to pay papers for online access (which I'm willing to do), we should expect good online practice to be followed. BTW, there is also no place for comments on this story in the online NYTimes, so I can't complain there.

Subsequent information:

Good and Bad News about the Safety of Natural Gas Fracking

Fractals Goes Social

If you wanna be read, you gotta be social. I've been meaning to properly support Facebook and Twitter on Fractals of Change for a while, but didn't get around to it until reader @tydanco chided me "why not stick a twitter button at the bottom of your blog so we can retweet it".

So I did. It's down at the bottom of each entry and you can not only tweet but also "like"a post on Facebook. If you click the tweet button, you get to modify a preformatted message which has the title of the post and a tiny URL for it. The tweet becomes part of the timeline of anyone following you on Twitter. If you like a post on Facebook, that get written to your wall so your Facebook friends can see what you like; Facebook may also post the event as news in your friends' news feeds but they are cagey about when they do and don't do this.

Twitter and Facebook also make it possible for people to find out when you post to your blog. It is this capability (plus email subscriptions) which kept my readership alive during the eighteen months I was working as Vermont Chief Recovery Officer and then CTO and only had time to post biweekly. People who used their browser to visit Fractals daily in the days when I did post almost every day quickly got discouraged and stopped coming when they rarely found anything new. But people who subscribe to Fractals via FeedBlitz email or follow me on Twitter or check my status on Facebook know when there is a new post and can read it if it sounds interesting. So, since I was already fooling around with layout, I decided to make it easier for people to use Facebook and Twitter to find out when there's a new post by putting the new stuff circled in yellow at the top of the left sidebar.

Clicking on the Twitter button goes to my @tevslin messages and makes is easy for someone to decide to follow me and get a notification in their own timeline whenever I post. The Facebook capability gives Fractals its own page on that service. People who "like" the blog by clicking on the thumbs up will get notified on Facebook when there are new posts (this post'll be the test. Hope it works). Previously only my Facebook friends saw new posts.

While working on all this, I accidentally discovered how important being social is. Readership has been down for the last month and I had no idea why. Turns out that about a month ago something broke on Twitter so that posts were not automatically becoming tweets and, since it was tweets that posted to Facebook, there was no notification on Facebook either. Looked like I'd gone radio silent. I think I'm now reconnected (again this post is the test).



Why We Can Succeed

After Governor Douglas' press conference last week, there was the usual scrum of reporters and administration officials in the hallway of the fifth floor of the Pavilion Building in Montpelier. The Governor had just said that Vermont cannot and must not sit around waiting for another Stimulus Bill (which would be a bad idea nationally) but can and must address its own structural problems in order to succeed in the post-bubble, post-recession (and post-stimulus) economy. In fact, he said, we are working towards a much more effective state government and have seen some early bipartisan support for some of the tough stuff we must do. Vermonters, he said, realize there's a problem, know there's no painless way out, and expect and will support action.

The reporters were, as reporters should be, skeptical. Is this just one more reorganization of government? If so, why do we think that it will be any more helpful than past reorganizations? Anyway, what are the specifics of the things we intend to do? What are these changes? What does a much more effective state government look like?

We administration officials didn't give the reporters the specifics they'd like to have – and I'm not going to do that here. Plans are not done; legislators not briefed; specific legislative proposals not written; and we have to propose a budget to deal with a anticipated gap of $150 million or more in fiscal year 2011 while figuring out how to permanently make state government more effective at lower cost to match revenues which will remain lower than in bubble years (and were being outstripped by expenses even then).

"So," a reporter asked me, "what is your role as Chief Technology Officer?" That is a question I can answer and helps answer why I believe we CAN succeed.

Part of my responsibility is to assure that we're using technology broadly-writ to make state government not only more efficient but also more effective. With a vast wave of retirements coming up in the state work force, there's an obvious opportunity to do the same work with less workers if we provide those workers the proper tools. But that's not enough; we want the work to be qualitatively better in terms of outcomes. Just for example: it's a good thing in itself if we can deliver needed benefits to needy beneficiaries with less overhead and fewer mistakes (efficiency); but it's even better if those benefits help more of the beneficiaries become independent (effectiveness).

Turns out that technology can be a big part of gaining both efficiency and effectiveness. Things are very different technically then when prior attempts at change fizzled or partially fizzled. We now have examples of how industries like airlines have used the web to dramatically change customer service both to reduce service costs AND to empower customers. Some readers may remember when you went to a travel agency to get paper tickets which were written by hand before you could fly anywhere.

Here's what's changed:

  1. The web exists and most people know how to use it to get service (but not all, of course).
  2. Even in Vermont broadband adoption is growing rapidly.
  3. The SmartGrid project by Vermont utilities and the VELCO fiber build (almost $200 million between them) will bring high-capacity data-carrying fiber into every corner of Vermont. This "backbone" capacity is necessary to get all our institutions online with very high bandwidth and to take us closer to meeting our residential availability goals even though the fiber itself won't stretch all the way to every home.
  4. We plan to up the goal from 100% broadband availability to 100% adoption (the Governor has proposed that over $3 million of remaining stimulus money be used for this rather than as a short-term bandaid). This assures that government services delivered electronically will be accessible to almost all Vermonters.
  5. The cost of computing resources continues to follow Moore's law down (50% reduction in cost for the same capability every 18 months) so that many computing and communication dependent projects have gone from impossibly expensive to very affordable just in the last five years.
  6. We can now afford to gather the data we can't afford to govern without (but, as of now, don't have). 

So, from my nerd's POV, the realm of the possible is greatly expanded, especially if we don't let ourselves be bound by current organizational constraints. We not only have the means to quickly implement 21st century government in Vermont, we are ready for change and all its attendant discomforts, confusion, and fears. 

  1. No alternatives to much more effective government other than a drastic cut in benefits to those who can't afford a drastic cut, raising taxes which will quickly get us less income as taxpayer flight accelerates, or the fantasy of more and more "stimulus" forever from a federal government whose credit is running out.
  2. A much better realization from at least some lawmakers of the situation we're in. The Governor talked in his press conference about the very real cooperation and good thinking coming from Joint Fiscal Chairman Michael Obuchowski and others.
  3. The acceptance (subject to ratification) of a paycut by state employees. That makes them part of the solution rather than part of the problem. We now owe it to them to give them better tools (technology, again).

It's going to be a very tough year in the legislature and in the state; but I'm optimistic that we'll be able to focus on the right issues, make the right choices, and do more than just "recover".

On The Front Page of the New York Times


Fractals of Change made the front page of the New York Times online edition, thanks to a robot of course.

The Times has a new feature called Times Extra which inserts hyperlinks to blogs which relate to front page stories in a scrollable box under the story as you can see above. I posted my criticism of Treasury for pre-announcing a plan for a subsidized interest rate cut and thus keeping the housing market in freefall until the cut materializes. The robot, which is part of a Times-owned service called blogrunner, recognized the relationship to the Times story (and was kind enough to disregard the fact that my post keyed off a Wall Street Journal article about the preannouncement).

Although, as a New Yorker, I was brought up to believe that being above the fold of the NYTimes is the road to fame and fortune, the extra traffic from this link is so far negligible – far less than when Techmeme picks up a post. This may be because Times Extra was just introduced yesterday and/or because you have to click the "Try Our Extra Home Page" button in the masthead below to get into Times Extra mode. Most visitors to the front page aren't in the habit of doing this.


But it's nice to be part of the conversation.

We Are ALL Part of the Change We Need

Tim O'Reilly blogged yesterday on the need for those who supported Obama in the election to stay involved and help the new administration but, more importantly, the country succeed. Those of us who didn't support Obama have an equal obligation: we can't sulk; we can't work for the failure of the Obama administration; we have to work for the success of the country no matter who gets the credit if for it and regardless of future elections. To do that we have to work with the administration wherever possible and be constructive and principled when we disagree. Change IS needed.

Tim asks "What do we do next?" and gives four substantive answers and one procedural.


  1. "Actually apply for one of the jobs in the new administration." (This certainly applies even to those who didn't work for getting the administration elected. We should take the new administration at its word that it'll be open to former opponents and new ideas.)
  2. "Whether inside or out, the tech community can continue to lead by example."
  3. "Identifying specific proposals for best practices and points of leverage."
  4. "We really need to weigh in on the issues that matter. From climate change, to open spectrum, to education policy, to investments in science and technology, we need to make our voices heard." (These voices, even from the tech community, won't be unanimous but they need to be heard. Particularly easy for us bloggers to do.)

Procedurally Tim proposes posting ideas to the transition team's website change.gov. Tim does note that the site itself should change so that submissions to it are not just one-way emails but visible to other visitors who can then comment. In the spirit of constructive suggestions,  I'd suggest that anyone have an ability to post to it – subject to after-the-fact moderation for egregious abuse. Also would strongly suggest that the blog on the change.gov site be open to comments. That would be a mechanism for and a demonstration of the change we need.

DISQUS for discussions

Lately the comments and replies on Fractals of Change have gotten a lot more intense and interesting, maybe because these are interesting times.

Typepad comment handling is pretty primitive so I've switched to DISQUS starting with this post. DISQUS supports threaded comments, has a more powerful profile system for commenters across blogs which use DISQUS, lets you (and me) get email notification of replies to comments and even lets us reply to comments with comments by reply email. Should be neat; let me know what you think.

For testing purposes, I'll be making the first comment on this post.

Full disclosure: I'm an indirect investor in DISQUS through Union Square Ventures.

Blogging 101 – Comment Spam

Sometimes you will find comments on your blog which link back to commercial sites and are there to promote a product or boost the google-juice of the linked to site. Are these spam? Should you remove them all?

Here are two comments which were recently posted on Fractals of Change. IMHO, one is legit and one's not:

"what a nice post Good luck!!!!!!!!!!!!***[commercial name removed] Challenge"

"We have recently created a free link exchange directory http://www.link-exchange-submit.com. Providing people use it fairly, it should help you with your SEO link building. It is automatic in terms of your link and suggested category being added. But it is also moderated daily in terms of spam removal."

The first comment is clearly spam in that it contains no internal reference to the content of the post. For some reason, the people or bots that post these almost always stick in a generic compliment thinking, perhaps, that you are starved for praise and will leave the comment posted. This one was deleted and reported to Typepad.

Since the post that the second comment was attached to WAS about links, the comment is right on target (although it could have been posted by a smart bot). My post was against link exchanges and this is clearly a contrary view; but that's fine – good, in fact. The service which is being promoted is something a blogger thinking about link exchanges would want to look at. So the comment is likely to be useful to readers of the post. It gets to stay.

Similarly, a post of how to make cheaper cellphone calls abroad is festooned with comments containing links to SIM providers and cheap or free calling services. Again, anyone interested in the post is likely to be interested in the links. Sure, they serve a commercial purpose and I don't get paid for them as would get paid for ads. But the comments expand on the post and make it more useful to readers. On FOC, that's the spam/no spam criterion.

Blogging 101 – Exchanging Links

My friend Al is a new blogger. He's received his first requests for link exchanges and would like to know what I think of this practice. For non-bloggers: a link exchange is an agreement that says I'll put a link to your blog on my blog if you link to my blog from your blog.

The sites which have solicited Al do not particularly impress him; but they do have fairly high link ratings. That means that, if they point to his blog, he will gain Google juice and Technorati rank and possibly readers. All he has to do is point to them in exchange.

Should he agree to the proposed exchanges? If you're a blogger, you've probably already faced this question.

Links ARE important. I said that in the first post in this series and also described here how the math of a group of cooperating blogs gives the whole group a competitive advantage compared to blogs not in the group – an advantage that can be significant out in the attention-starved desert of the long tail. So should you or Al do link exchanges?

I don't do link exchanges because I want my readers to find that my links are relevant to them. I also don't because I want the blogs that I do link to know that I value them – but can't deny that I hope that they'll find that the value is mutual and that their readers may appreciate my posts and link to me when appropriate. So you might say that I do actually tacitly do link exchanges. I don't insist on a quid pro quo; I don't link to content which isn't either relevant to the current post or likely to be of general interest to my readers.

Readers who come through random links won't become regular readers; they have no reason to. Fractal of Change is eclectic (not a great way to build circulation – focus is better) so sometimes an automobile blog or a company blog or an environmental blog'll point to a single post of interest to their readers. FOC can get a flurry of hits this way and I'm glad for the chance to sound off on whatever the subject is to these people outside my regular reader group. But FOC doesn't get many return visits from these random hits – not enough here to interest them. On the other hand, if a VC blog or nerd blog or even an energy blog points here, some of the new readers become regulars because there's enough content relevant to them to keep them.

Assuming you want regular readers and not just more hits (which may not be true), then link exchanges aren't likely to get you there because they somewhat devalue your content. On the other hand, link exchanges with blogs much more linked to than yours WILL get you a higher Google rank for the subjects you write about than if you didn't have those links so you do pay a price for being a purist (nb. rumor has it that Google bots know how to devalue references from blogs which would otherwise gain authority from link exchanges but I haven't heard this directly from any bot's output device).

What IS a good strategy is to engage in a blog to blog dialog including comments on each other's posts with a group of blogs which cover the same subject matter as your blog – better if you're all NOT always in agreement; better if the links to each other have more substance than "Joe just posted something great"; but links which get each other readers AND give readers value by advancing a discussion are the best kind.

BTW, linking to new blogs which catch your interest is a social obligation of existing blogs,

Should Your New Web Business be Ad-Supported?

Contrary to popular belief, the ad-supported model for a web business is very, very hard to succeed at. I've underestimated the difficulty of living on ads a couple of times in the my investment career. Wouldn't want you to do the same either as an investor or an entrepreneur.

Google, of course, is advertising-supported and is a huge success; that's bad news for you - and Yahoo and Microsoft. Google is consuming most of the free oxygen in the ad-supported cave. If someone wants to buy keyword-driven ad inventory, they go to Google. Why should they bother going anywhere else, especially to a startup very few page views? If you want ads on your service, you'll have to sell them or get them from someone else who is selling them. No matter how neat your self-help ad engine is, no one except maybe your mother will try using it UNLESS you open the way to a significant new market AND can prove that.

OK, you say, people won't by ads from me but they'll buy them from Google. I'll create the page views with my wonderful new service and Google'll put the ads there. Google'll put ads anywhere. Yeah, but. Google will put ads on your pages; some people will click on these ads (Google charges for and pays for clicks in case you've been living on Mars and missed that). You will get a small stream of revenue; it won't bring you anywhere near breakeven. It won't impress potential investors. In fact, the trickle of revenue you get from Google might even convince potential investors that you CAN'T make a living with ads; no ads and no revenue might leave them easier to convince.

Google ads are fine for harvesting ADDITIONAL revenue. Bloggers run them because any revenue is nice; but most bloggers aren't trying to make a living from their blog or attract investors to it. If your website sells something, it makes all kinds of sense to sell additional related somethings through Google ads or Amazon ads (which you can better aim at your customers). If your website has some spare space, Google ads are something people are used to looking at and they'll make some spare change for you. But they won't make your business model.

Maybe there's a counter-example (if so, please post it). In theory, I thought, since Google does such a good job of keyword targeting and increasingly good job of geo-targeting, a well-defined site that viewers come to for well-defined reasons (not an eclectic site like Fractals of Change) ought to be able to induce the Google bot to send just the right ads to attract many clicks at a high price per click. But I haven't seen it happen that way.

Besides Google there are ad networks which will actually sell your site to advertisers. Professional blogs do get a great deal of their advertising from networks like Federated Media (FOC is a small blog using FM). But it's tough to get the attention of a good ad network if you don't already have good demographics AND high viewership. Even if a network takes you on, their salesmen aren't going to be able to do much for you unless you have numbers big enough to get their attention and the attention of the advertisers and agencies they sell to. If your content is powerful enough, you might make a living with agency ads – but it's a long shot if you're not BoingBoing.

Whether it's Google or an ad network, whoever sells your ads is going to have to keep a lot of the revenue to pay the selling expense. It's highly unlikely that they'll be enough left for you to run your business on.

The bottom line is that you have to have way to sell ads if you're going to support a service on advertising revenue. Sell as in actually convince somebody to buy something, not just take orders. Selling ads nationally means having existing contacts with people who buy ads nationally AND having such a hot property that they'll pay attention to you. Selling locally means feet on the street walking into stores and helping to build local campaigns. Radio stations know how to do that; you probably don't and probably can't afford to hire someone who does. The easy local ads were the classifieds because the newspapers didn't sell them, they just took orders. Craig and his List jumped quickly into that huge niche. Almost inconceivable that you'll make money there.

If you have a better way to sell ads, then maybe you should start an ad-supported service. If you just have a better service, you probably can't support it on ad revenue alone.

Too Much Revenue, Not Enough Growth

Jeff Jarvis posted a comment on my post In Praise of Revenue:

"I'd also like to see you reprise your lesson (from the Union Square event some time ago) on extracting minimal value from the network you create so the network grows as large as possible and the value you've created and can extract in the end is greater than if you had tried to extract more value at the beginning. Did I get that right? I quote you to that effect all the time. Did it just the other day with a big publisher whose blog ad network is taking too high a cut. I told him to just cover his costs for the first year - or less - and he'd end up growing something bigger that would be more valuable to each member, thus bigger, thus more valuable to him. Eh?"

Jeff remembers quite accurately that I advocated optimizing for growth rather than revenue – in the extreme forgoing revenue. Jeff's advice to the publisher was right on. If you simply solve for maximizing revenue, you can end up with little growth – and little future revenue opportunity. Note, though, that Jeff did not advocate forgoing revenue; in fact, he did advise the publisher to cover costs, presumably so that growth can occur without needing to raise more capital or so that there will be a solid basis for raising capital when it can be put to good use.

The case Jeff presents of an ad network is particularly straight forward. It is difficult if not impossible to sell ads which will be seen by only a few number of people. The cost of selling the ads is too high to justify the effort; advertisers are not interested in taking the trouble to investigate a tiny potential market or put any creativity into reaching it. Other than in strictly local markets, there need to be millions upon millions of impressions AND data to do targeting with before advertisers are interested. So it is not practical for any but the very largest blogs to sell their own ad space – and even they usually don't. There is an opportunity for ad networks which aggregate advertisers and advertising on one side and an inventory of space ads can run on the other side. The network matches the ads to the blogs, typically collects from the advertiser, and pays the blogger. Google is the most successful example of an ad network but the ads it aggregates appear in many more places than just blogs. Federated Media, the ad network to which Fractals of Change belongs, is an example of a blog-based ad network.

If you're an ad network, the more page views you have to sell, the more and better the advertisers you can attract. The more advertisers and the higher the rate for page views you can achieve, the more bloggers you'll attract to make their page view inventory available through you. You obviously have to scratch to get started, need to have some credibility or an existing inventory of ads to start with, and are going to lose some money getting going. But now you've got traction: how much of the ad revenue should you share with the bloggers and how much should you keep? Your investors may be pushing for some return on their capital (profits); your compensation might even be tied to your margin on sales rather than just your gross sales. Nevertheless, charging more than you have to, even if you can for a while, is a mistake.

Charging too much stunts growth so you'll have fewer units to charge for in the future. Charging too much opens the door to competition.

The more that bloggers make from your ads, the more space for ads you'll have available as bloggers tell their friends which ad network to use. The more ad space you have, the more ads you'll get and – on the average – the more you'll be able to charge for ads because you'll have better opportunities to target and you'll have more advertisers interested. The more ads you get and the more you can charge for them, the more money bloggers in your network can make.  You want to keep this virtuous circle of growth going as long as you possibly can.

If you are extracting profits before you have to, you're forgoing future growth. In any sort of competitive market, profits attract competitors. Big profits attract lots of competitors. Would-be competitors can point to your profits and easily get funding. Funded competitors can undercut your rates and "steal" your bloggers. Whoops; the circle is now turning in the non-virtuous direction. If you're doing well but running at or close to breakeven, you've made it impossible for anybody to undercut you without running at a deficit which is hard to get funding for – at least in this market. The biggest danger to you is someone who finds a way to substantially cut costs or to deliver a better product. Obviously you've got to be vigilant about that and ought to lose some sleep over these possibilities – but keeping prices down keeps a plague of me-too competitors from cutting off your growth.

This logic goes well beyond ad networks, they just make a good example.

Craig's List has the successful strategy of forgoing revenue for MOST listings it runs and MOST markets that it's in. That strategy helped it attract a critical mass of listings and a critical mass of listings meant a critical mass of ad readers which attracted more ads etc. etc. If Craig now attempted to maximize revenue by charging for a substantially higher percentage of ads, a door would be cracked open for competition. There is no chance at current rates for a competitor to steal Craig's listings (and readers) by charging less. If and when Craig's List is bested, it'll really have to be by something which delivers a better way for listers and readers to communicate.

Unless you are a protected monopoly, high prices are a recipe for losing whatever lead in the market place you have. Low prices are the engine of growth.

The strategy Jeff suggested to the publisher and that I'm recommending here is to keep revenue as low as it can be and still fund growth. No revenue is a different strategy that I'll post more about.

Now on Kindle!

hackoff.com: An historic murder mystery set in the Internet bubble and rubble

CEO Tom Evslin's insider account of the Internet bubble and its aftermath. "This novel is a surveillance video of the seeds of the current economic collapse."

The Interpreter's Tale

Hacker Dom Montain is in Barcelona in Evslin's Kindle-edition long short story. Why? and why are the pickpockets stealing mobile phones?

Need A Kindle?

Kindle: Amazon's Wireless Reading Device

Not quite as good as a real book IMHO but a lot lighter than a trip worth of books. Also better than a cell phone for mobile web access - and that's free!

Recent Reads - Click title to order from Amazon


  • adlinks
  • adsense