The Problem with relying on Google
I've used Google's search engine and most of its products for a good few years now. The clean interface and the search results themselves initially drew me in, having previously used Altavista.
Since I studied GCSE IT, I've had a hotmail.com email account - but after receiving an invite to GMail from a fellow Star Wars Galaxies player, my use of hotmail started to decline. My domain email is all sent to my own mailserver, powered by hMailServer but is then downloaded by GMail periodically, as I couldn't find any free and open-source web mail interface that would rival GMail's. Since then 14,000+ emails have made their way into my Inbox. GMail revolutionsed email - no longer did you have to label and organise your email into sub-folders or delete them altogther. It's simply a case of read the email and archive it - letting Google's powerful search functionality retrieve it at a later date, if required.
Bookmarks are handled by Google too - I tried a PHP/MySQL self-hosted solution, but found the UI to be too clunky and the project appeared to be abandoned. With bookmarks, I like the way that Google integrates it into the search results - so that you can instantly see those you've previously starred.
As for Calendar, I have a few entries - but have never been much of a user. The same can be said for Docs.
Then there's Reader. It's a slightly different story - I have 117 subscriptions and often stumble across another RSS feed to subscribe to. I find it an addiction, checking feeds several times a day to ensure I'm up-to-date on the latest.
I also have data in Google Analytics for 10 domains; 1 purchase using Google Checkout; Chrome is synced to my Google Account; Feedburner has burnt 5 of my feeds; and Web History has proved useful in the past - as a better alternative to a Browser's own History.
Then Google went and launched its own Social Network, Google+
I'll admit that I don't use it as much as twitter or facebook - I've found each serves its own purposes. I read somewhere that facebook is for keeping in touch with the people you know, but don't like and twitter is for following the people you don't know but do like. Something along those lines. There's some truth in that - depending on who you follow/are friends with. I find facebook a way of keeping up with people I know: what they're up to and twitter as a way of hearing about the latest before anyone else. Google+ has carved out its own sector - detailed, intelligent technical discussion. Perhaps it's because the majority of the users on there are geeks at the moment, but I find each post on my Stream to be worth reading - all signal, no noise - whereas twitter is more noise than signal, requiring effort to find the gems.
Google+ is here to stay and with more people joining each day, it'll soon rival twitter and facebook for market share.
Then Google went and released Google Music. It wasn't long ago that I was looking for a cloud based music provider, that I could store my 1000's of MP3's on to listen to whilst at work. Previously, I relied on FTP to my server at home, to grab albums I wanted - and then import them into iTunes. This isn't too bad, unless you consider how bad iTunes has become over the years - Apple surprises me with each update, as the software appears to be worsening. As soon as I saw a tweet asking if anyone wanted an invite - I jumped at the chance. During registration, I was told that the Beta was currently US only - but was pleased to find that it worked for me in the UK, using a UK ISP.
I have around 120GB of music on my server here, but after installing Google's Music manager app, to begin uploading MP3s to the service, I started ignoring the shite and picking out the best albums instead. It was refreshing to find that I only actually liked a small portion of the collection - 4374 songs to be precise. Uploading it didn't take long either, thanks to a 50MB down/5MB up Virgin Media connection. The downside is that work's Internet connection is ~4MB at best, shared amongst 25+ users - roughly 164KB each if you think about it. This equates to a stuttery experience, which has meant that I barely use the service. Still, at least I know my music is there, waiting to be played.
I probably use a few more services that Google offers, but the above is the majority of my Google time. However, today I realised that relying on one provider for so many services can be a problem.
Take, for example, the Google Translate API. We use the translateth.is plugin to provide an easy interface into the API for Users to translate websites into their desired language. Google recently decided that people were abusing the API so much, that they are removing it completely by December 1st 2011. And it's not the only Google product to be axed.
Google recently announced that they're restructuring and focusing their priorities on 3 distinct areas of the business. Off the top of my head:
- Search
- Advertising
- High traffice services
Search is a given, as it is the core of their business. Advertising is where most of their revenue comes from - which leaves the high traffice services. Anything that doesn't fit into these 3 key areas is being discontinued. APIs that 3rd party developers and businesses rely on; Services that many found essential in their daily lives; and Google Labs - the birthplace of many of these new ideas.
From a business point of view, it makes a lot of sense - they've spent years having fun, creating what they could and testing different markets with new services. But they've now decided to focus on what makes them money.
From a user's point of view, it's worrying. Each time I hear of a new round of cuts - I wonder if a service I count on is going to be discontinued. They've spent time switching the high traffic ones over to the new Google+ style interface, but haven't yet touched upon Bookmarks or Reader to name a few.
There's a saying - don't put all your eggs in one basket. I've pretty much done that with Google.
I recently closed my Yahoo, Flickr and Hotmail accounts - instead using Google Picasa and GMail as alternatives. Today I logged into GMail to find an email from Google Apps, to say that my account had been updated to support all of the non-Apps services. I had setup Google Apps when it was first opened to the public, with the intention being that I would use it for all @slickhouse.com email. However, I didn't like the idea of changing MX Records to point to Google, when GMail was still in its infancy - and the fact that I couldn't use all of Google's products with the same Apps account.
However, as part of the update, Google removed the association of my @slickhouse.com email from my @gmail.com account and placed it with the Google Apps account, without prompting me. After reading the email and realising the problem, I immediately signed into the Apps account and deleted it. The trouble is, the deletion process takes up to 5 days and I cannot re-associate my @slickhouse.com email to my @gmail.com account. It's a minor annoyance, but suggests the whole process and system is flakey - with so many services, Google has created a fragmented account/login solution, which they're only now trying to piece back together.
I'm confident I'll get it back to how I want it - no more stale Apps account, but a single @gmail.com account with my @slickhouse.com email associated with it.
Therein lies another problem - my @gmail.com address was setup in a rush as I wanted to use the invite and try out this new GMail thing. As my domain email was forwarded to it, I never really noticed it - until Google+ came along. Google doesn't offer a way of associating more than one @gmail.com address to another - one feature that hotmail/live does allow for. So if I do decide that I can no longer live with my current @gmail.com address, I'm stuck with having to manually export my data from each of the services that allow it and then import it into the new account. Not all services allow this though - Google+ being one of them.
Maybe this is all the result of Google developing lots of small projects and purchasing startups, without any uniform way of tying them altogether. Now they've realised that they have too many products and not enough focus to move forward - that they're starting to claw back and tighten up integration between each of the products.
I'm not against Google in anyway - I'd even go as far as saying I'm a fan. But I've come to realise that I shouldn't be relying on them for everything - even though their products are reliable, fast and intuitive. At least as a Web Developer, when the time comes - I can develop my own alternative(s).
StackOverflow and ServerFault
For those of you regularly turning to Google for answers to your programming woes, you'll find StackOverflow to be essential in day-to-day development. In fact, due to the popularity of the relatively new site, you've most likely come across many questions and corresponding answers within the Google results - often ranking very high.
The site itself is very clean and easy to use and after a few visits you'll soon pickup how it all works. I've been a member for just over 4 months, but have only clocked up a reputation of 121 - as I'm more of a reader than contributor.
On the flipside, there is now ServerFault from the creators of StackOverflow - Joel and Jeff. It's geared towards the system administrators out there, to discuss server memory configurations and connecting to SQL Server 2008. As I take a keen interest in the hardware side of things too, ServerFault is a great read for me too, but my reputation is even lower.
So, if you're not already using StackOverflow and/or ServerFault then start navigating there now. If not, you'll be seeing more of them both as the content from the community grows and is crawled by search engines more and more.
Posted: Thursday 18th June 2009, 05:54pm
Categories: Computers and Technology, Links, Web
Tags: google, serverfault, stackoverflow
Google in 2001
As part of Google's 10th Birthday celebrations, they've re-instated their oldest search index from 2001.
It's interesting to see how both search engine algorithms and the internet has evolved in the few years since.
So far, I've found that I didn't exist on the Internet in 2001, facebook was unheard of and wikipedia didn't feature as highly in the search results as today.
Posted: Wednesday 1st October 2008, 05:55pm
Categories: Computers and Technology, Links, Web
Tags: google, internet
Google Chrome
Google's open source Chrome browser has been released!
Naturally, being a web developer I was itching to get my hands on the new browser, mainly to see how it performs compared to the other popular browsers available and more importantly to see if my sites worked well enough to not warrant more browser snagging.
Although the browser is in the beta stage (isn't everything these days?), its performance exceeds IE7, Opera and Firefox on many tasks. Javascript is the most noticable improvement - with Google Mail and Reader zipping along 50+ times faster than IE7.
Chrome does have a few flaws, to be expected with a new release - such as the lack of a Google Toolbar, poor accessibility and numerous bugs already spotted by bloggers across the 'net. But, its simplicity and forward thinking more than makes up for it.
Something that Mozilla, Microsoft and the likes are probably kicking themselves about right now, is that Google's Chrome handles tabs independantly - each within its own process. For those of us using 10's (even 100's) of tabs simultaneously, we'll appreciate this feature the most. Ever had numerous pages open at once, only to have one of the tabs crash on you, bringing down the whole browser as it does so? Chrome solves this along with other single-threaded/process issues.
Time will tell if Chrome will build up a user-base to rival IE7, Firefox, Opera and Safari. I'm currently using it alongside Firefox/IE7 at home and may end up replacing the latter permanently if Chrome meets all my browsing needs.
For other developers out there, you'll also be pleased to hear that it uses an existing rendering engine behind the scenes - WebKit (as used by Apple's Safari), which should mean that if you've already checked your site(s) in Safari, then you shouldn't need to alter much else to be Chrome compliant.
Let me know if you're trying out Chrome and if you'll be using it long-term...
Posted: Thursday 4th September 2008, 05:55pm
Categories: Computers and Technology, Web, Work
Tags: browser, chrome, google
Updating WordPress Permalinks
After ironing out all of the bugs with my latest theme release, I decided to update the permalinks used throughout slickhouse.com
Thanks to WordPress' intuitive admin interface, this is a simple process:
Most WordPress users opt for the Day and name Permalink structure, which would appear as 2008/05/04 for today. However, to a search-engine spider, this already looks like the links is 3 directories deep, before it even gets to the post title itself.
To counter this, I've amended the permalink structure to mirror the ISO 8601 standard for date (minus the time) - so today reads as 2008-05-04. To a search-engine spider, this would appear as just the one directory, within which the post title is situated.
There is a downside to amending permalinks to a WordPress blog that has been live for several months (even years) - all of the current links will effectively be broken. You could use a redirect script, that detects the user is trying to visit an older page - which would redirect them with a 301 (permanently moved) redirect to the amended url. However, this would still mean that your older links are present in your data.
To resolve this, you can amend all occurences of the older permalink to the new format. So, you'd look for 2008/05/04 and replace it with 2008-05-04.
Using MySQL Administrator Query Browser, you can query the WordPress database, as above. The first step is to ensure you're querying the correct database (essential if you have multiple databases on the one server):
Use [database-name]
Where [database-name] is the name of your WordPress database. A standard install usually prefixes the database name with wp_. Execute the above and you'll see the correct database highlighted in the right-pane. Next, you want to begin searching for posts which contain the old permalinks within the content:
Select ID, post_title From wp_posts Where post_content Like 'example.com/2008/%' Order By ID Desc
If your WordPress database is fairly small (<500 posts), then you should find that the above query executes almost instantly - leaving you with a list of all posts that use the older permalink structure. You'll need to replace example.com with your domain name and run the query several times - once for each year your WordPress blog has been active, amending the year each time (2007, 2006 etc.).
Once you've got your list, you can start editing the posts to update the permalinks used. This could be achieved using SQL, but I would recommend doing it post-by-post to save any headaches if your update query goes horribly wrong!
Using the WordPress admin (Manage > Post) you can edit each post using the HTML view:
After editing a few, you'll soon quickly spot the a tags, ready to update them. In my case, it was simply a case of replacing the / with -, so 2008/05/04 became 2008-05-04.
There was a reason the SQL query brought back both the ID and post_title - as you'll soon find out when trawling through the admin. The admin orders (by default) the posts by ID, which means ordering the results of the Select query should make finding the offending posts a lot easier. To make things even simpler, you can amend the URL each time you have finished editing the one post and are moving onto the next:
As you'll see - the post.php references the posts by their IDs.
Once you've updated both your permalink structure and your content of each post, you'll be left with a fully working site. However, you may also want to notify the major search engines of your recent update. For this, I can highly recommend the XML Sitemap plugin for WordPress. After setting it up, it will generate an XML Sitemap of your WordPress blog and simultaneously notify each of the search engines of the updated sitemap.
Then, it's just a case of sitting back and awaiting the spiders as they begin to crawl your new, friendlier (in theory) permalink structure. If you haven't already setup Google's Webmaster Tools, then now is a good time - as it will help diagnose any issues with your new permalink structure.
Finally, if you don't have direct access to your WordPress database, then you could try using the built-in search facility, as seen in use on many themes. Simply search for your older permalink structure and it should disply all relevant posts.
Posted: Sunday 4th May 2008, 05:55pm
Categories: Articles, Articles, Slickhouse, Web
Tags: google, permalinks, sitemap, wordpress
Feeds
Twitter
