When Cuil launched just a couple of months back on the 28th July, it jumped right in there with the ridiculous claim that it was already the biggest search engine on the planet right from the get go. This was, apparently, based upon some clever math which ended up with Cuil indexing some 120 billion web pages, or more than three times as many as anyone else. Even Google, which at around the same time stuck a tentative claim in for a trillion unique web pages identified during the automatic index creation process had to admit it only actually indexed a percentage of those in order to weed out the duplicates and dynamically generated content.

What Cuil did was mix claims of the biggest web index with claims of unique content-based relevance methods for the engine, and promised to go beyond link analysis and traffic ranking in order to provide quite simply ‘the ideal search engine.’

Sounds impressive, huh? And the fact that Cuil was the brainchild of former Google search index architect Anna Patterson and Stanford University search research pioneer Tom Costello could not hurt either. “By leveraging our expertise in search architecture and relevance methods, we’ve built a more efficient yet richer search engine from the ground up. The Internet has grown and we think it’s time search did too” Patterson exclaimed at the time.

Certainly that seemed to be the case at first, the publicity worked and the promise of a better-than-Google-search added up to a very quick carving out of a 0.26 percent share of the global search market. OK, so not something that would have Google executives quaking in their boots, nor shake the foundations of SEO the world over, but nonetheless impressive enough for a first 48 hours figure.

Yet Net Applications, the monitoring and measuring company which came up with those numbers has been keeping an eye on Cuil and now reports that things are not so hot, in fact they are positively icy cold. According to the latest figures for the end of August the Cuil market share has dropped to, well, just about nothing at all. Net Applications has the market share being a very meagre 0.01 percent. At least it is a very steady 0.01 percent though.

Why is this? Well I think I might know the answer: when compared to Google, Cuil search results are crap. A big steaming pile of crap in fact. I put it to a real world test after launch, going head to head against Google across a number of really very simple and very real world searches. The results were not just disappointing, they were devastating as far as Cuil was concerned. Other reviewers, bloggers, analysts and reporters were busy doing much the same and reaching much the same conclusions. That the technology behind Cuil might be impressive, the search index might be massive, the pedigree of the developers faultless, but the results were just not good enough. Certainly not good enough to switch from Google, not good enough to tarnish the Google brand one little bit in fact.

And those Net Applications market share numbers would seem to suggest that end users agree.

Not that everything is rosy for Google, because Net Applications has also announced a rather dramatic slide in popularity of the open source Google Chrome web browser. Chrome enjoyed a respectable one percent global market share within the first 48 hours, an unprecedented statistic. Now, though, it has slid down first to 0.85 percent last week and just 0.77 percent this. It seems like Chrome has been tarnished by one too many incidents involving security and stupid copyright grabs but above all else just does not deliver on the smaller, faster, safer promise. End users are, it seems, reverting back to Firefox and Internet Explorer, although Safari would appear to be doing best of all out of the Chrome migration.

It has been a long while coming, but following the acquisition of Postini by Google last year the enterprise version of Google Apps is finally getting the security it deserves and its users demanded. The new Google Web Security for Enterprise is now available as a web service and incorporates real-time malware protection along with policy enforcement and URL filtering.

Tim Johnson, a Google Product Marketing Manager, has blogged about how it enables “the safe, productive use of the web, without incurring hardware, upfront capital, or IT management costs” although I am not sure where that left enterprises who were using Google Apps for their business remotely up until now. The free add-on extends the kind of security that in-house workers enjoyed to off-network ones as well, which is of course good news albeit bloody late in the day if you ask me.

“Protecting off-network users used to require them to connect via a VPN when they were out of the office — often with mixed results” Johnson admits “With this new feature, all off-network users’ web traffic is automatically directed to scanning infrastructure to enforce your policies and protect their computers, requiring no action on the part of individuals. It’s easy to deploy and users can’t tamper with it, ensuring your security and appropriate use policies are always in place.”

I guess it was only a matter of time: before Microsoft made a real effort to buy Yahoo, and once it did before Google started stamping feet and shouting that it just isn’t fair. Considering that Google has something like 80% of the search market as far as the UK is concerned at least, compared with around 10% combined for Microsoft and Yahoo, it does rather stick in my craw when it starts complaining about ‘unacceptably dominant positioning’ to be honest. Yet that is exactly what Google is doing, warning anyone within earshot that if Microsoft buys Yahoo then it will create a dominating email and instant messaging monster which could jeopardise future development of the open standards Internet.

Of course, similar concerns have not come to the fore when Google itself has been on the acquisition trail to strengthen its position as a provider of online services. Of course, it is just the kind of puff and bluster to add fuel to the fire after the US justice department announced it would investigate (for antitrust reasons) any deal between the two online giants.

If you ask me it just confirms that Google is worried that Microhoo could become the first serious competition to its own position in the marketplace, the online advertising marketplace that is. Funnily enough, that is one area of unacceptably dominant positioning that Google has been suspiciously quiet about…

Talk about social networking sites and you probably think Facebook, MySpace and possibly LinkedIn. The chances are, unless you happen to be Brazilian, the Google social networking offering Orkut has managed to evade your radar altogether. Orkut is, however, hugely popular in Brazil and that’s why it was mostly Brazilians who were among the estimated 750,000 members to find themselves on the wrong end of a worm infection within the space of 24 hours.

According to McAfee Avert Labs the Orkut worm, which spread rapidly at the end of last week, was highly targeted towards that Brazilian community. It used the Orkut scrapbook facility, a method of sending messages between ‘friends’ in the network, to distribute text scraps in Portuguese which translated into such cobblers as “2008 is coming, I wish that it begins quite well for you.” It also made sure that an executable was downloaded which added users to an Orkut community group called “Infectados pelo Virus do Orkut” which translates to “Those Infected by the Orkut Virus.” The loop continues and the worm spreads by then sending more infected scraps to everyone within that persons network of friends.

It appears that all of this was made possible by a relatively new feature of an Orkut tool that allows its members to write message scraps containing HTML code, the new feature adding Flash and Javascript capability into the mix. Uh oh, the warning lights and sirens should have been going off when news of this ‘functionality’ was released.

It all meant that the only action needed to be taken for someone to get infected was to view their Orkut profile. Which seems to be the point, as the group description reveals that the worm was actually created and released to expose just how dangerous the system is.

Certainly the worm caught the attention of Google quickly enough, as it would seeing as it was spreading at the rate of 100 member infections every minute at one point and it would appear code filters have been put in place to prevent further exploitations.

“Google takes the security of our users very seriously. We worked quickly to implement a fix for the issue recently reported in Orkut. We also took steps to help prevent similar problems in the future. Service to Orkut was not disrupted during this time” a Google spokesperson said.

Now I have never encountered a CAPTCHA entry box while searching at Google myself, but apparently a number of people have. My mailbox has had a number of queries from concerned readers of various publications I contribute to, including IT Pro, asking just what the heck is going on.

Good question, so I thought I would do a bit of digging and find out.

Most people who have contacted me are concerned with one of two things:

1.       The inconvenience factor of adding precious seconds to a quick search routine, enough so to be considering moving to another search engine seeing as the market isn’t exactly short of decent ones.

2.       The implication that their security is pants because the CAPTCHA warning screen says that ‘a computer virus of spyware application is sending us automated requests and it appears your computer or network has been infected.’

It is the second of these points that concerns me most as the last thing we need is someone like Google spreading FUD (Fear, Uncertainty, Doubt) about such things for no good reason. Oh, and the reason that Google is throwing the CAPTCHA’s at ya, well that’s simply because it is countering the automated bot search flooding that it experiences pretty much on a daily basis. Trouble is, innocent parties are getting caught in the crossfire when their ISP, or rather the ISP’s proxy server, is used by the bad guys.

Of course that does rather get Google off the hook and put your ISP firmly on it. If someone is abusing Google and using your ISP to do it, then can you really blame Google for taking action? Far better, surely, to throw up the CAPTCHA system than just blacklist all requests and present you with a 404 of some sort.

So my advice to anyone seeing the dreaded Google CAPTCHA screen on a regular basis, as after all a single occurrence could be a blip, is to start chasing your ISP to track down and terminate the offending user. If they can get any kind of dialogue going with Google that is, although that’s another blog posting story altogether.

I also have a little advice for Google, which I hope will excuse me for being so presumptuous. Can you not map offending netblocks to owners rather than simply assume all connections from that ISP IP are problematical? While I can only start to imagine the problems you must face from the bot herding scum out there, pushing those problems on to innocent end users is not exactly great when it comes to relationship building and customer retention.

Google has just confirmed its third largest purchase behind DoubleClick and YouTube, and it points towards a consolidation of its efforts to tackle Microsoft head-on in the business applications market. The £310 million ($625 million) price has been paid for corporate security specialists Postini. With a particular emphasis on email monitoring, data encryption and compliance, the Postini acquisition would seem to be set to equalise things following the Microsoft purchase of FrontBridge a couple of years back.

Certainly it will help to strengthen the corporate case for Google Apps Premier Edition, which apparently has already garnered 100,000 small-business sector users according to Business Week.

But what does this all mean as far as the email security marketplace is concerned, and will Google succeed in making Gmail and its assorted other apps more competitive within the corporate sphere? Those were some of the questions I asked of Craig Whitney, Managing Director EMEA at Tumbleweed Communications, another big player in the messaging security market. Here’s what he said, in full:

“Google’s acquisition highlights its efforts to compete directly with Microsoft – by acquiring Postini, Google is providing virus and spam protection for its Gmail users, and will also likely leverage Postini’s user base. This follow’s Microsoft’s purchase of FrontBridge two years ago, which provided similar email security to Hotmail. Postini has approximately 10-11 million individual users – primarily individuals and small and home-based businesses. However, this move does not get them into the coveted enterprise space that Microsoft has infiltrated. It also remains to be seen how Postini customers will respond to Google’s policy of scanning emails for advertising.

This move is significant, however, as it highlight the competition between the two companies and underscores the importance of the messaging security industry. The segment is growing rapidly and there is tremendous potential — companies like Google, Cisco and Microsoft see this opportunity. Meanwhile, enterprise security tends to lead the charge in innovations to messaging security – five years ago, it was only large enterprises that were focused on this, and now individuals and small companies with less than 250 employees are looking for solutions. Today, messaging security at the enterprise is not about spam and virus protection alone. There is a proliferation of ‘content in motion’ over the internet, and preventing data leaks and managing large file transfers are two major capabilities that are essential to larger enterprises. Overtime, we see these as becoming central to businesses of all sizes – it’s only a matter of time.”

It could be game over as far as the Google Gmail web-based email service is concerned in Germany. Not only has the battle between German venture capitalist Daniel Giersch and Google over the Gmail trademark been settled in favour of Giersch, but the final nail in the coffin as far as Google is concerned might well be a new German law concerning the retention of personal data.

This is not the first time that Google has faced a trademark battle and lost in Europe. Here in the UK it was forced to change from Gmail to Googlemail for that very reason. The data retention law is a different matter though, and really does look like making it a tough call for Google to stay and play with the Germans.

The problem is that this law would seem to require Google to not only obtain the real names and addresses of its Gmail users, but verify them as being genuine as well. Gmail has built its success largely upon the relative ease of getting an account, and the fact that it is anonymous helps this immensely.

According to a news article on the German language Heise Online news site, Google’s German spokesman Peter Fleischer has made it clear that this anonymity is key to its continued success, stating that “if the web community won’t trust us with handling their data with great care, we’ll go down in no time.”

Looks like that time is now, or at least very shortly then, as far as Germany is concerned.

The rumours surrounding a Google OS just will not die, fuelled by so called ‘leaked’ screenshots of GooOS that are, in fact, nothing more than a mocked up version of Gnome with some custom wallpaper. The chances of Google actually developing some kind of custom operating system for anything but internal use are, it seems to me, extremely low.

An online experience is one thing, but moving this offline into the very different world of computing platforms is something else. Who would be the target market for a start? What business would see the value in switching to a Google based OS over and above any number of Linux options or Mac OS X if considering switching at all? The same argument does not apply to the more easily impressed consumer of course, and the brand attraction of Google coupled to the availability and ease of use of the growing number of Google apps could be a compelling one. Some kind of WinCE-alike mobile device pre-loaded with the Google stable running under a Google branded OS could feasibly have quite a ready made market.

But would the GooOS be a truly new, innovative and ground up build if Google did prove the doubters wrong and please the pundits, well this pundit anyway, with such a thing? The answer is, I suspect, no. The front-runner would have to be something based on Ubuntu, already being referred to as Goobuntu, as this is apparently being used by the search and marketing giants internally at the moment. It has even been reported that the Google derived bugfixes and improvements are being shared with Ubuntu for integration into the core distribution.

Personally, and this is the geek in me speaking I readily admit, I would love to see something like this make it out into the real world just to see how long-trousered Google could make it without submitting to the urge to create some toy OS. But I really am struggling to think of anything that Google might do to differentiate it from the competition.

So, here’s a question for you: if you were in charge of developing a Google OS, what would you do?

Look, I can understand that any business with a lucrative revenue stream from a resource with a distinctive look and feel is going to want to prevent ‘passing off’ in any shape or form. That is just common sense and business sense merging into one. What I cannot understand is when a company imposes restrictive terms upon its customers, upon the very people that allow it to function, in order to not only prevent such passing off but also prevent the use of any competing service. Such contracts are usually illegal, are they not?

Anyhow, that would appear to be exactly what Google is doing with the newly announced changes to the AdSense policy on ‘competitive ads and services’ which in effect lays claim to ownership of your website design if you want to continue to enjoy whatever tiny slice of the advertising revenue pie serving Google advertising brings. The relevant change to policy reads:

“In order to prevent user confusion, we do not permit Google ads or search boxes to be published on websites that also contain other ads or services formatted to use the same layout and colors as the Google ads or search boxes on that site. Although you may sell ads directly on your site, it is your responsibility to ensure these ads cannot be confused with Google ads.”

Whatever happened to Do No Evil, huh Google? Whatever happened to wanting to be different, not only in how you approach search but also how you approach business in general? Whatever happened to not being Microsoft?

Already pretty much the most influential single corporate in the world of online advertising, and please don’t fool yourself that Google is in the business of search because that became secondary to smelling the advertising dollar a long time ago, it looks like Google now wants to up the ante and go the whole hog towards monopolisation as far as the lucrative contextual advertising market is concerned. Some might argue, and I am close to being one of them, that Google also wants to extend that reach into how you design your website and how you deliver your content.

Will complaining make any difference? Hell no, Google is too big and too self-important to worry about a little negative blogging or even a few hundred folk pulling their AdSense accounts. There will be  hundreds of thousands more who will adopt the mantra from the early days of the mainframe computer admin “BOHICA.” That’s ‘Bend Over, Here It Comes Again’ for the whippersnappers who might want to remember it for the next Microsoft move Google makes.

In fairness, there are plenty of exclusions in the policy both explicit and implied, and there is not a total ban on all contextual ads. Indeed, you can still run other adverts on the same page as AdSense ones. What Google will not allow are ads that ‘resemble’ AdSense, at least not on the same page. Which is all well and good, but the grey areas in this policy are such that I am concerned that it will become a matter of interpretation and Google will, as always, be the judge and jury. Confusion is never a good thing when it comes to policy, and even an increasingly lumbering corporate leviathan like Google should understand that. What about where ad images are lined up alongside AdSense ad text? Surely a design matter for the website owner? Well Google would rather you didn’t do this as idiots, sorry I mean users, might think the two were related and click on the wrong one – which would be the non-Google one from their perspective one assumes. Unless the dimensions of the image are different to the dimensions of the AdSense text box, then it is OK. See what I mean about confusion and design interference?

So Google has, as expected, announced a number of new features for the Google Search Appliance, the enterprise level search engine in a box that is actually a rebadged Dell PowerEdge with Google inside along with the dual-core Intel Xenons. So what is new, and is it exciting? Well, the new stuff consists of the following, and I’ll leave it up to you to decide just how exciting each feature is:

Results Hit Clustering –
This enables users to drill down on any given subject, effectively and dynamically creating results based subcategories for better search refinement. The ability to customise both location and appearance will, I am sure, be welcomed by most admins as will the time this functionality should shave off of separating the relevant results from the piffle.

Source Biasing –
This menu driven tool makes it relatively easy to assign relevance to search results based upon the authority of the content source, as it impacts upon your business. The fact that this does away with the scripting requirement often found for such functionality in some old school enterprise strength search solutions is laudable.

Query Expansion in Additional Languages –
Not before time Google has introduced some extra languages by way of query stemming and synonym dictionaries for French, German, Italian, Portuguese and Spanish. Hurray, or should that be che cosa ha preso così lungamente?

Sitemap building –
The ability to automatically build a sitemap for Google.com’s crawler is now featured, making your public website content more readily available to the Google search bots.

Open Source Connectors for Microsoft SharePoint –
If indexing and searching content that lives within Microsoft SharePoint 2003 or 2007 is a critical requirement for your enterprise, then Google has woken up to that fact and now embraces your needs. Oh lawdy yes.

And least but definitely not last, Enhanced Security Options –
Well, support for NTLM version 2.0 authentication/authorisation anyway.

The added functionality goes some way to offset the criticism levelled at Google for not enabling the search relevancy algorithms to be ‘tuned’ to best suit individual, and ever changing, enterprise search needs. Whether the make it easy options added are enough to satisfy those needs, of whether ultimately enterprise clients are happy to continue with the manual programming slog in the quest for perfect results, remains to be seen.