Monday, October 31, 2005

Undercover

I've been on SEO forums and newsgroups since the day dot. I remember the glory days and although I find most forums to be clique filled speculation pits of ego boosting trash - I don't regret the time I spent on them in the early days.

I am glad that I've kept, more or less, undercover. I give up a forum alias if it becomes too widely known and move on. I just like to keep my head down - which is why it's taken so long for me to start this blog, why I'm not promoting it and why it's covered with disclaimers and caveats.

I had to abandon one of my aliases as searchenginewatch's forums (which are, I think, probably the best forums today) as it picked up a whole whack of reputation. I'd been posting with this alias for months, generally seeking out those posts with no or only a few replies and providing as helpful an answer as I could. You don't win reputation for anything like that. I made the mistake of linking to an updated page within Google's guidelines and that was enough to spark around of votes and "keep up the good work" comments (which grates against my arrogance as they sounds patronising ).

My current alias is about as easy to guess as could be so I imagine I'll abandon it in a little while - or perhaps put it out to stud, perhaps for return visits when I want to make a post in a semi-anonymous way.

On the topic of undercover forum use we have Matt Cutts and GoogleGuy. For the longest time the wide spread belief was that Matt Cutts was GoogleGuy. This is still likely to be the case. However, Cutts has begun to cite GoogleGuy posts in "almost" the third person over on his blog. He'll say something like, "Over at forum X, GoogleGuy has posted to say ... ". Of course, this does not rule out that Matt Cutts is GoogleGuy. He could be referring to the 'forum alias GoogleGuy has posted' and still be grammatically correct. Is it 'evil# to lead us to believe something by implication? A little. Whereas Matt might have been safer to leave well alone, I can see that he might also want to draw as much attention to GoogleGuy posts as possible in those situations where he wants people to read them and his blog is a good vehicle for that.

Saturday, October 29, 2005

WebProNews

Arhg! It's articles like this one at WebProNews which really annoy me. In some ways it's better than many WebProNews articles because it's not full of grammatical and spelling errors (which are fine in a blog but not in a professional publication). That said, this is still a dangerous piece of twaddle; rumour and speculation are presented as absolute truth and fact.

In this article the author states that Google punishes you if you buy text link links. Your competitors can go out and buy text links "on your behalf" to get you punished.

I wonder how he can square it against Matt Cutt's own explanation of Google's views and actions. It's the link selling site which is effected once Google cracks down. Google removes that site's ability to pass PageRank on. Sure, if you had text links from a site which suddenly lost it's ability to vote for you then you will take a dip in the SERPs. That's not quite the same thing as your own site being punished.

That said there is a time when a competitor could cause you trouble by buying you links; when your site is brand new. That is likely to cause your site to stand out (it'll be an outlier) and your site is much more likely to what the SEO industry has coined the sandbox effect.

Arhg! Do you know what currently comes top for that sandbox effect Google search? You guessed it - some useless speculation about what the Sandbox Effect is on WebProNews.

I have a small confession to make though. The other weekend I wrote an article for WebProNews about Search Commands. Nothing dramatic. The old joke about the Med, the broken Convert command, etc... it's not appeared. Guess I wasn't sexy or speculative enough to cut the harsh and very strict WebProNews editorial rules!

Friday, October 28, 2005

Yahoo! Search blog: Video Searching: Now Easier Than Ever!

Yahoo does seem to have the lead in video search. Yahoo's video search makes it easy to have your QuickTime and media files included and it covers all sorts of movie media on the web. Google's Video search, on the other hand, seems to only really bother with TV media and capturing it. I suspect Google could extend the sitemap XML to include specialties like video or audio. The tweaks to Yahoo Video are responsible for a big announcement over at Yahoo! Search blog but they are just tweaks. The API news is bigger and better.


Just wait a second. The other, earlier, entry into the Yahoo! Search blog is, I think, much much better. They've finally added a "Save to My Web" button and even gone a step further to let users search their tag clouds.

I think this really will help to promote Yahoo's Social Search activities. It was the orange RSS and "Add to MyWeb" buttons which started to spread throughout the blogosphere which really did push RSS from fringe to mainstream. I suspect the same will happen here again even though the My Web button is quite bulky. You'd be a die hard user to put the tag cloud on your web site but only last night I was toying with parsing the RSS feeds from my Social Search at Yahoo as a form of a "recommended links" page for ARHG net. I may just slap on the tag cloud instead.

Thursday, October 27, 2005

Google Travel


Yahoo is all over travel; they have the new Trip Planner in Yahoo! Travel and they bought Farechase. Since Yahoo! announced Trip Planner today the naughty among us might have expected Google to do something to divert some attention back their way.

It so happens eagle eyed Google users caught the search engine testing a new travel GUI. I've stolen this image from Search Engine Lowdown.

So what's going on here? Google's spotted a trend in the keyword search - that someone is considering going from A (Atlanta) to B (Madrid) and responded by inserting a specialised search box above the SERPs. We've the choices of Expedia, Hotwire and Orbitz. This is controversial - imagine you're a big travel company and you're not there. This is also very American; over here we don't have Hotwire or Orbitz. For example, Lastminute or Thomson aren't represented here. The lack of Thomson would be more significant as they're a huge travel company and do not operate through the likes of Expedia. Another tricky example would be something like the easyGroup who have a number of companies but not in one site, there's easyJet, easyCar and easyCruise to name but a few.

Google has a mixed track record here. Let's leap in and talk about toolbar three as it introduced the auto-mapping and auto-isbn features.

Auto-mapping allowed uses to press a button on the toolbar and get links to maps automatically added to the web page they're currently on in suitable places. Google had Google Maps at the time and has Google Local (the two merged) so they could have exclusively used Google's own maps but at the minute you can pick a different provider if you want. Sure, the selection is not huge but despite Google having a commerical incentive to push their maps there is some wriggle room here.

The auto linking of ISBN numbers is not so good. Press the button on the toolbar and you'll have any ISBN numbers on the page linked to Amazon. It's a monopoly despite other online bookshops being available. Certainly here in the UK Tesco Books have as wider and often cheaper selection.

There might be a circular situation too. If Google continues to improve it's results through personalisation and, at the same time, promotes the current market leaders then Google's users are going to show a "personal bias" towards these market leaders and so Google's personalisation will promote them further. If, for example, this travel GUI links to British Airways Holidays for UK users (if they make the distinction) then a lot of Google users are going to search and then click through to BA Holidays. This could bias their search results, through personalisation, in favour of BA Holiday

Wednesday, October 26, 2005

Thrilldare?

Is www.thrilldare.com a Google domain? It shows a copy of Google currently.

Big deal. Lots of sites soft redirect to Google or just clone a few pages.

ping www.thrilldare.com and ping base.google.com - most of the time you'll get 64.233.187.99.


Pinging base.l.google.com [64.233.187.99] with 32 bytes of data:
Reply from 64.233.187.99: bytes=32 time=122ms TTL=241
Pinging thrilldare.com [64.233.187.99] with 32 bytes of data:
Reply from 64.233.187.99: bytes=32 time=142ms TTL=241


Even so, that could just be an akamai thing. However, a tracert does firmly put the IP address in Google's hands.

Google Base

There is no such thing as Google Base. Yet.

We do have a rather compelling screen shot and a Google Blog post about the latest rumours.

I think it is likely that Google Base (and the name could change) is real. This is something Google would like to do. Google wants to index the world's information and a large community database is one good way to do it. There are pros and cons with a community database; as Wikipedia knows, a successful community database expands rapidly with a wealth of information but there is an issue as to whether the data is accurate. A wiki is good here as anyone can correct wrong information. A database with a single point of entry for each field is not likely to have that advantage. Google already has Google Answers where experts are paid to answer questions. This system has the benefit that the "good enough for the customer" answer is highlighted as accepted so other viewers can see the data had some worth. The screen shot of Google Base suggests you could add a "Database of protein structures" (this is a typically Googlesq example) and it certainly leaves me wondering how I can tell an accurate protein structure from one made up. There's the same problem on the main Google index, some could argue, but on the Web no page is in isolation. A page claiming to discuss protein structures and which is linked to by trusted authorities is more likely to be correct than a mess of a page which no one links to (that's how PageRank works, after all).

Classic. Here we find a blog posts which goes from discussing whether there might be - could be - will be a Google Base and concludes by discussing the pros and cons. We're still to see whether there will be a Google Base and we're certainly still to see how it works.

There...

Having said that I doubted if I had time to update this blog often enough - I seem to be writing another entry.

I've sorted out the atom.xml feed. It's being published to ARHG Net correctly now. Phew.

Atom is an interesting one. It's there because RSS is too 'owned'. Atom's certainly more powerful but it's also a faff. Even when Google seemed so reluctant to go down the RSS route, RSS feeds were clearly the syndication option worth recommending. Yahoo had embraced RSS and most of the traffic driving syndication software had too. Further more, RSS is easier for affiliates to get their head around and therefore easier for them to hook into and use. Now Google seems to have some more time for RSS; you can add it to your version of the Google home page and you can get RSS content out of Google via Google News. The .rss file extension is still shown as unrecognized though and that's just Google being slow on purpose.

Although RSS is ahead of Atom in popularity this position is not sacrosanct. If RSS's 'owners' do try and through their weight around then I suspect Atom would simply scoop defectors up by the bucket load. I remember how many webmasters dumped their .gifs in favour of .jpg when the courts ruled .gif was patented.

Tuesday, October 25, 2005

There we have it

I thought I'd give UKShells a go, though I am yet to muck around and see whether I can have command line access to my web areas (for mysqldump commands or wiki installs, say). I actually tried to do this last night but it proved impossible to register new domains at set up because of a bug in their system. I rang sales today and was impressed; no hard sell, just friendly and fast action. I even picked up a token discount for my bug spotting.

So, we have www.arhg.net with no content yet and www.arhg.net/blog/ for the blog. I know. How original. ARHG may be my initials but the temptation to type ARGH is still there.

Hosting companies are a strange breed. They're an example of one of those internet companies which require a mass of customers to turn a tidy profit but who begin to hemorrhage profit if they begin to get bogged down in support. I have accounts (my first) at UK Linux and at Fasthosts. I asked Blogger.com and Fasthosts support the same question at the same time, Fasthosts replied over night and I'm still waiting on Blogger.com. Fasthosts support do get knocked awfully but the company does look after an incredible number of web sites. One day I may even have an account at RackSpace UK and I suspect it's most likely to be GameWyrd although I'll be pleased, in a way, if it turns out to be one of the new ones.

A start

You have to make a start somewhere. For me this start was October the 25th in 2005. It was a day where I've dashed back and forth between Hillside and Portobello as I attempted to move, piece by piece, the contents of my flat.

But wait. No. This is not that sort of blog. I have one of those already, it lives on LiveJournal where I think blogs like that belong.

This blog should be a different animal. I work in search engine optimisation, a form of internet marketing, search engine marketing if you would like (and some people do prefer the term). Each day I try and find enough time to check what the popular search blogs are saying, I'm checking for official news from the search engines who maintain blogs themselves or what the industry gurus think. I suppose we're lucky that we have an industry where the gurus are so loud on their blogs, in this case blogging and opinion go hand in hand with the industry landscape. We still influence each other.

I am lucky. I work for one of the best SEO firms in the world. No doubt of that. At times I find it ironic that I'm so busy I rarely get to read what other blogs say. I might be being a bit daft in thinking that I have the time to write a blog of my own. I'll try. I want to. From time to time I just feel that an industry or search engine development needs a comment. Whereas this blog is not the place I spleen daily foo, this blog will be the personal-professional vent for search engine and search engine optimisation news and views.

Just as I find it ironic that some people seem to have 16 hours a day to crawl the forums and update their blogs I also find it worrying at how often I disagree with what's been said and how often cliques seem to perpetuate status and standing. I suspect I'll skirt with controversy here.

That said, let's get this important disclaimer in; this blog represents my views and not the views of the company I work for. Personal views, such as those you'll find in this blog, change, evolve, twist and change. I reserve the right to flip-flop, u-turn, reserve or bathe in sea changes.

I'm also new to Blogger. Here's my first discovery; the Firefox version of Google's toolbar does not spellcheck properly in the "Edit HTML" of the Posting screen. The input window turns blue but no errors are found and you need to click the ABC button to disable the effect (rather than being able to mouse-click on the form and select "Stop" from there.

Here's the next discovery: if you start spell checking in "Compose" view, flick to "Edit Html" without turning the spellcheck off then the Html inherits all the style commands from the toolbar, going back to "Compose" hardwires these changes into the actual blog post.

The next step for me is to move the blog to a new domain.