It's not very often that I talk about search engine marketing anymore. It used to be a good deal of what I would do and write about. The last three months I have been doing more of it as many of the projects for Thalasar come to fruition. These means I have been spending time slogging it out checking out what has changed over the last few years. The funny thing is, the more things change, the more they stay the same. One thing that hasn't changed is the often overlooked directories. One of my favorites is a directory called, Best of the Web!. It originally started in 1994 around the same time as Yahoo!. They did alright in the boom years selling basic banner ads. They always had a mix of eclectic and original web sites in the directory. As time went on they didn't become Yahoo but they didn't become Starting Point either. They muddled along and they have experienced a new resurgence as a usual eclectic place to find things.
With so much time being spent on SEO and link building campaigns, Best of the Web! is a great place to start. With a broad directory you will certainly find a place for your site. I highly recommend Best of the Web! as directory you should definitely use in your marketing efforts.
Monday, March 28, 2005
Monday, March 07, 2005
Evolutionary Momentum , Software Design & Dislocation
I was in Pasadena visiting with Andrew Maltsev of Ejelta to discuss design of the new web forum object to Xao Software. Of course you might be thinking,"Yet another forum? Isn't phpNuke enough?" Well yes and no. In designing the forum object, we took a close look at what everyone else was doing. Remarkable most of the forum software out there looked remarkably similar. It was as though every developer that decided to develop a forum package did the exact same survey and wrote their feature set based on it and called it a day.
This seems to the common way to write software today. Open Office & WordPerfect are perfect examples of trailing the leader, MS Office. In a way this make it easy to set your feature set and decided your development path. The problem is that as a result software that fills a specific niche begins to look the same, and act the same. This can be a benefit as similar user interfaces allow users to easily use either application. Of course this doesn't really lead to new radical feature set development but something just close enough to be familar and different enough to be frustrating.
So does design dictate software features or does the software function in our version of "nature vs.nuture." The answer is very clearly both. Forums look a certain way because that's the way that people expect forums to look. Slashdot which offers a great deal of personalization of the software for each individual user. Most everyone I know who visits Slashdot views it in nearly the same way. I think the effort customization takes exceeds it's minor benefits. Clearly software in a specific niche is going to look the other critters inhabiting the same part of the software ecosystem. However just because there are certain functional requirements that dictate design, doesn't mean we should stop there. Most of the time however, that is exactly what happens. However many software packages are designed by following a leader in the market and hoping that me too will minimize adaption headaches. So once an evolutionary niche is carved out by a piece of software, evolutionary momentum drags everyone along it's path.
Of course it doesn't have to be this way in design. After Andrew and I discussed the required features based on our observation of competing products, we immediately launched into a discussion of what we disliked about their products and what sorts of shortcomings they have. This then lead into some pretty innovative feature ideas. One lead to an entire new business idea (currently under development). So just matching a current offering is one thing, finding it's flaws another.
This is why so often dislocating technologies do not come from established industry players. Established players such as Microsoft Office are too far wrapped up in their model to think radically. Since such dislocation often threatens an existing revenue stream or entrenched development team so they creep along slowly and stealthly until some notices, "We aren't selling as many licenses as we did this same quarter last year." Eventually this new technology stealths in like a small mammal and completely dislocates and destroys most of the previous market. Of course part of the market will still exist dinosaur like for a number of years.
I think we beginning to see the some of the dislocating technology right now for the traditional office suite, namely a suite of web enabled tools such as wiki, and web based calendering. Over time these will likely evolve into a useful replacement for Office and "Office Like" technologies. Office and it's clones are mainly about document production. While it's useful to produce a nice looking document, it's much better to actually product useful work. In a networked economy, more network centric apps are more useful. It's still the early stages though and MS might adjust. I think its much more likely that that this sort of growth will occur underneath MS's radar. MS major issue with Office is that no seems to be upgrading. This is hardly surprising as far as I can tell there is no major difference between Office 97 and Office 2003. So their interest is on selling upgrades as opposed to inventing the next paradigm for collaborative work. Indeed inventing the next paradigm will require them to "knife the baby" and potentially destroy an existing revenue stream.
This seems to the common way to write software today. Open Office & WordPerfect are perfect examples of trailing the leader, MS Office. In a way this make it easy to set your feature set and decided your development path. The problem is that as a result software that fills a specific niche begins to look the same, and act the same. This can be a benefit as similar user interfaces allow users to easily use either application. Of course this doesn't really lead to new radical feature set development but something just close enough to be familar and different enough to be frustrating.
So does design dictate software features or does the software function in our version of "nature vs.nuture." The answer is very clearly both. Forums look a certain way because that's the way that people expect forums to look. Slashdot which offers a great deal of personalization of the software for each individual user. Most everyone I know who visits Slashdot views it in nearly the same way. I think the effort customization takes exceeds it's minor benefits. Clearly software in a specific niche is going to look the other critters inhabiting the same part of the software ecosystem. However just because there are certain functional requirements that dictate design, doesn't mean we should stop there. Most of the time however, that is exactly what happens. However many software packages are designed by following a leader in the market and hoping that me too will minimize adaption headaches. So once an evolutionary niche is carved out by a piece of software, evolutionary momentum drags everyone along it's path.
Of course it doesn't have to be this way in design. After Andrew and I discussed the required features based on our observation of competing products, we immediately launched into a discussion of what we disliked about their products and what sorts of shortcomings they have. This then lead into some pretty innovative feature ideas. One lead to an entire new business idea (currently under development). So just matching a current offering is one thing, finding it's flaws another.
This is why so often dislocating technologies do not come from established industry players. Established players such as Microsoft Office are too far wrapped up in their model to think radically. Since such dislocation often threatens an existing revenue stream or entrenched development team so they creep along slowly and stealthly until some notices, "We aren't selling as many licenses as we did this same quarter last year." Eventually this new technology stealths in like a small mammal and completely dislocates and destroys most of the previous market. Of course part of the market will still exist dinosaur like for a number of years.
I think we beginning to see the some of the dislocating technology right now for the traditional office suite, namely a suite of web enabled tools such as wiki, and web based calendering. Over time these will likely evolve into a useful replacement for Office and "Office Like" technologies. Office and it's clones are mainly about document production. While it's useful to produce a nice looking document, it's much better to actually product useful work. In a networked economy, more network centric apps are more useful. It's still the early stages though and MS might adjust. I think its much more likely that that this sort of growth will occur underneath MS's radar. MS major issue with Office is that no seems to be upgrading. This is hardly surprising as far as I can tell there is no major difference between Office 97 and Office 2003. So their interest is on selling upgrades as opposed to inventing the next paradigm for collaborative work. Indeed inventing the next paradigm will require them to "knife the baby" and potentially destroy an existing revenue stream.
Thursday, February 03, 2005
"Descarte Limit"
Time for me to add a new term to the lexicon. As this will certainly come up in the next thirty years, I have coined the neologism "Descarte Limit" as it it applies to AI.
"Descarte Limit" (Noun) The Descarte Limit is the limit below self awareness for an AI. An AI below the Descarte Limit would lack self awareness. Once an AI has passed the Descarte Limit, it has become a self aware entity. Signs of an AI having passed the limit would be awareness of itself outside its normal operating parameters, awareness of events and systems outside its normal operating parameters and a reflection of its role in its operating parameters.
For an AI passing the Descarte Limit is the initial step for full recognition as a sentient being.
As a legal term the Descarte Limit is the litmus test for an AI seeking legal protections or recognition as a legal entity.
Taken from the 17th century French Philosopher Rene Descarte famous principle, cogito ergo sum ("I think, therefore I am").
I will probably add this to the wikipedia.org as well.
"Descarte Limit" (Noun) The Descarte Limit is the limit below self awareness for an AI. An AI below the Descarte Limit would lack self awareness. Once an AI has passed the Descarte Limit, it has become a self aware entity. Signs of an AI having passed the limit would be awareness of itself outside its normal operating parameters, awareness of events and systems outside its normal operating parameters and a reflection of its role in its operating parameters.
For an AI passing the Descarte Limit is the initial step for full recognition as a sentient being.
As a legal term the Descarte Limit is the litmus test for an AI seeking legal protections or recognition as a legal entity.
Taken from the 17th century French Philosopher Rene Descarte famous principle, cogito ergo sum ("I think, therefore I am").
I will probably add this to the wikipedia.org as well.
Tuesday, February 01, 2005
Longhorn Canceled?
Fuckd Company is running a rumor that Microsft will be canceling Longhorn. While I highly doubt this, I suspect that the completely rebuild the operating system, (which is what Longhorn was originally supposed to be has proven to be quite a difficult and daunting process. Reinventing instead of re-factoring is always a mistake in approaching a software project. This simply means you will make a different set of mistakes this time around and introduce new bugs. This assumes of course that you have developers which have developed an operating system before - widely recognized as one of the most difficult things to do. The GNU project struggled for years to develop their micro kernel, HURD, which is is still not ready for prime time. I suspect that a more incremental approach is going to win out at Microsoft which will make Longhorn like Windows 98 was to 95 - a temporary upgrade to be followed by a new version in 2008 or so.
Sunday, January 23, 2005
Longhorn - Feature Phishing Nightmare
The feature set of Longhorn has been reduced several times. To my mind this is a good thing as it prevents more buggy software from entering the Windows ecosystem. WinFS is a by all internal reports, not ready for prime time as it is the first pass by MS. One continuing problem is that Microsoft's annoying and dangerous tendency to integrate new features into the OS that serve business purposes rather than technical purposes. The most horrific example is of course Internet Explorer which resulted in thousands of viruses being spread at the cost of billions of dollars to corporations. To date no one has really held Microsoft responsible for the enourmous costs they forced on the world in their efforts to destroy Netscape. Every decision has unforeseen consequences but in the case the consequeces where pretty obvious. Of course Microsoft is protected by their EULA which means they do not bear the financial cost of their huge mistake. I mention this as an example as MS is about to do it again. This time with the new and improved desktop search integrated into Longhorn. Once again here's chairmen Bill Gates from the same article I quoted the other day.
The first error that Mr. Gates makes is mistakenly equating his companies search product with Google. As my last post pretty clearly demonstrates, Microsoft has a LOOONG way to got to even match Google's current offering.
The compelling business case for this integration that it allows MS wrap the "search" experience from files to the Internet into a MS branded experience. This will allow them to "choke off Google's air supply," by owning the entire search experience. Really this is about ending a competitive threat and much less about. While there might be some debate about this. Some people may think I have got far too much crap on my hard drive. I need MS to help find stuff. For those people I recommend X1
Of course the real problem is the unintended consequences of this integration. While a technical case could be made for improving finding files - it's not really a hugely compelling case feature wise. The unintended consequence is that is provides crackers and other associated eavil doers with the capability to quickly find what they are looking for in a cracked server. Now passwords, credit card files, documents and anything you want to keep private will be quickly found. The feature will undoubtedly ship active from MS which means it provides a universal access point for all data on a machine. MS will undoubtedly bundle a server component which will also prove open to attack and rooting the machine and thus all machines on the network.Every time I hear integration from Microsoft, I think ok, how many viruses/root exploits will developed because of this integrated "feature."
Time will tell if I am right but for now I recommend actually buying X1. It's fast, secure and is much better than either Google or MS for desktop search.
". . .-our research agenda will allow us to take today's search from ourselves and Google, and make what we have today look like a joke. And a lot of that will be built into applications like Office or the Windows shell. I see our desktop search offering--I think every review I've seen has rated it far better than what Google is coming out with.
When you get to Longhorn, it will even have deeper integration, and we'll have the same index format. So anybody who wants a smooth transition to Longhorn where you don't have dual indexers and everything--the commitment we're making at our desktop search is that same indexing, same format, and we'll make that very smooth for people."
The first error that Mr. Gates makes is mistakenly equating his companies search product with Google. As my last post pretty clearly demonstrates, Microsoft has a LOOONG way to got to even match Google's current offering.
The compelling business case for this integration that it allows MS wrap the "search" experience from files to the Internet into a MS branded experience. This will allow them to "choke off Google's air supply," by owning the entire search experience. Really this is about ending a competitive threat and much less about. While there might be some debate about this. Some people may think I have got far too much crap on my hard drive. I need MS to help find stuff. For those people I recommend X1
Of course the real problem is the unintended consequences of this integration. While a technical case could be made for improving finding files - it's not really a hugely compelling case feature wise. The unintended consequence is that is provides crackers and other associated eavil doers with the capability to quickly find what they are looking for in a cracked server. Now passwords, credit card files, documents and anything you want to keep private will be quickly found. The feature will undoubtedly ship active from MS which means it provides a universal access point for all data on a machine. MS will undoubtedly bundle a server component which will also prove open to attack and rooting the machine and thus all machines on the network.Every time I hear integration from Microsoft, I think ok, how many viruses/root exploits will developed because of this integrated "feature."
Time will tell if I am right but for now I recommend actually buying X1. It's fast, secure and is much better than either Google or MS for desktop search.
Saturday, January 22, 2005
Bill's Gates' Search Comments
In this same article Bill Gates also made some interesting comments about search. I have linked to the pagewhere he begins to talk about search. I wanted to take a look at these as they are very revealing about Microsofts's plans and more importantly failed thinking about search.
Here's one quote,
This is quite true. MSN search has been around for years, powered by Inktomi, with ads from Overture. Despite being in the search business before Google they have not made a single innovation in web search. That said Bill thinks the business opportunity here is in "search." It's not. It's in "finding." It's about finding the right information. It's about finding the right business. It's about finding the right quote. But that's search in another word you might say. What search is really about is information and providing it within a contextual framework that makes sense. Google knows this which is why local search works at Google. That's why Google's digital library is so important. Google knows that being the gateway to reference material and trustworthy information is the future of search. MS doesn't. If they did, they would roll out Encarta as free resource. Instead they are keeping Encarta gated behind a subscription model with 4,500 articles for free. This makes Encarta much less useful than the Wikipedia. You cannot share the information (unless of course they have a Encarta subscription). Google realizes that web search is the end to the means of finding the right information - NOT finding the right web page.
Here's a money quote
This is a money quote because it shows Bill is certainly (eating his own dog food/drinking his own Kool Aid) because local search works perfectly at Google. Let's try a test case. I live in Las Vegas, NM. It's a small town with a population of roughly 19,000.
So I want to find the local pizza place at Google I type the following;
"Las Vegas NM pizza"
or
"pizza las vegas nm"
or
"pizza las vegas new mexico"
and I find all three pizza places in their local results section at the very top of the page with their addresses and phone numbers. This is because Google has thoughtfully bought a database from Axciom which contains all local United States businesses and preloaded their database.
When I try that at the MSN preview site I get total garbage. Result #1 is pretty close at USdiners.com but I actually have to go to the web site to actaully see. The next two results are for Las Vegas, NV. This despite the fact that I explicitly told the search engine where I was located. For those of you who want to try it here's a Las Vegas NM pizza on MSN search
What's even more interesting that there is a "Were you looking for 'pizza' near Las, Vegas, NM" at the top of the page. I think ok, this should enable me to narrow down the selections as it really understands that query. Clicking this link instead shortens the search term to just pizza so they can show me ads that they are hoping I click on. The #1 organic result is the OahuWiki followed by free pizza notice at UNM. Keep working MS. Realize that search is just about finding the information you need, not searching web pages. I was going to tackle search embedded with Longhorn, but this post is getting somewhat long. Soon I promise.
Here's one quote,
"We've been in the search business even before Google was around"
This is quite true. MSN search has been around for years, powered by Inktomi, with ads from Overture. Despite being in the search business before Google they have not made a single innovation in web search. That said Bill thinks the business opportunity here is in "search." It's not. It's in "finding." It's about finding the right information. It's about finding the right business. It's about finding the right quote. But that's search in another word you might say. What search is really about is information and providing it within a contextual framework that makes sense. Google knows this which is why local search works at Google. That's why Google's digital library is so important. Google knows that being the gateway to reference material and trustworthy information is the future of search. MS doesn't. If they did, they would roll out Encarta as free resource. Instead they are keeping Encarta gated behind a subscription model with 4,500 articles for free. This makes Encarta much less useful than the Wikipedia. You cannot share the information (unless of course they have a Encarta subscription). Google realizes that web search is the end to the means of finding the right information - NOT finding the right web page.
Here's a money quote
Oh, sure, everybody is working on those things, but just take the idea of finding your local pizza place and doing that right; search doesn't do that well today. Search is really crummy today--it's just that it used to be really crummy, and now it's better, and there never was anything like this before. So most of the results people get back today are irrelevant results. Deep analysis can take us much further, and that's why we're investing a lot, and you'll see us more(sic) very rapidly.
This is a money quote because it shows Bill is certainly (eating his own dog food/drinking his own Kool Aid) because local search works perfectly at Google. Let's try a test case. I live in Las Vegas, NM. It's a small town with a population of roughly 19,000.
So I want to find the local pizza place at Google I type the following;
"Las Vegas NM pizza"
or
"pizza las vegas nm"
or
"pizza las vegas new mexico"
and I find all three pizza places in their local results section at the very top of the page with their addresses and phone numbers. This is because Google has thoughtfully bought a database from Axciom which contains all local United States businesses and preloaded their database.
When I try that at the MSN preview site I get total garbage. Result #1 is pretty close at USdiners.com but I actually have to go to the web site to actaully see. The next two results are for Las Vegas, NV. This despite the fact that I explicitly told the search engine where I was located. For those of you who want to try it here's a Las Vegas NM pizza on MSN search
What's even more interesting that there is a "Were you looking for 'pizza' near Las, Vegas, NM" at the top of the page. I think ok, this should enable me to narrow down the selections as it really understands that query. Clicking this link instead shortens the search term to just pizza so they can show me ads that they are hoping I click on. The #1 organic result is the OahuWiki followed by free pizza notice at UNM. Keep working MS. Realize that search is just about finding the information you need, not searching web pages. I was going to tackle search embedded with Longhorn, but this post is getting somewhat long. Soon I promise.
Subscribe to:
Posts (Atom)