Peter's Blog

A comment made to this blog led me to finding some gems for my new phone:

Google Mobile Maps

A java application that does google maps! I don't have a sat-nav as I would only need it about twice a year and it's cheaper to get lost. I usually print a google map before I leave home and make sure I have memorised enough to understand what I am doing.

I'll probably still do that but having google maps in my pocket will put my mind at ease when venturing into unfamiliar territory. I always get panicky when my fuel gets low and I have to find a diesel outlet. These days they tend to be hidden away at the nearest backstreet Tesco rather than being on the main roads.

The application includes a primitive (accurate to about 1km) location finder! And it can show satellite photos(!).

Opera Mini

A much nicer browser than the built in browser. It shows the whole page and you can move around it easily and zoom into the bit you want to read. This terse description doesn't do it justice. Very nice, especially in landscape mode.

I'm hoping the novely of my phone will wear off soon as otherwise I will need a data plan. I've spent £3.60 so far and each google map page is 80k of data.

Google Notebook

Not a java app but a link to the mobile version of Google Notebook. Post notes from your mobile and access them from your desktop without messing with syncing/email.

The mobile internet on my new phone seems much faster and more usable than it did on my old phone. The main thing that helps is that the phone goes straight to a google search page whereas on my old phone I had to wade through a load of vodafone branding. I am finding it handy to be able to google whereever I happen to be rather than have to remember to do it later.

I think the last time I used the internet on my old mobile was when I had sold my car and needed a taxi home from the wilds of Thurrock, alien territory. I went through the vodafone crap and found a search page and searched for local taxis. I got about three references, one didn't answer, another was limosine hire and the third was a bus company. I walked for miles that day.

I think (the phone doesn't make it obvious) I am using plain old WAP for browsing. I don't think any gprs is set up and I've turned off the 3G search as I have read that trying (and continually failing) to scan for a 3G network is a waste of battery power. The performance on plain old WAP is ok.

At work one of my chores is to set up the email on the managers new smartphones. That job has put me off the hassle of having one of those and the speed that the email downloads at isn't wonderful. I have installed the gmail reader on my mobile and that was much easier than trying to set up imap on a nokia. Oh, if only the company used google apps...

I received an email from google asking me to confirm I wanted to reset the password on one of my gmail accounts. It's as if some swine is after one of my three gmail addresses. The one in question is my main personal account and has 6136 messages in it, consuming 388Mb of my 6440Mb allowance (I only delete spam). My two other gmail accounts (petersblog and one with a sensible name for 'business') feed into this account.

I love gmail and don't want to mess around with any desktop email applications. That was until now. Do I trust google security? Do I want to have to archive it myself? Another job for my slice?

I can't resist posting this. The mail headers check out and I don't think it is a wind-up. There are no non-disclosure terms so here it goes:

Hi Peter-

My name is {snip}; I'm a talent scout for the engineering team here at Google. I came across some of your work and your writing on the Web and was impressed with your development experience - particularly your background in Linux and your work in Python.

I actually recruit for the Google.com engineering team here, which is the group of engineers that's essentially the mission control of Google. They're responsible for the design and development of the infrastructure for all our web applications and internal services. You'd be able to tackle some of the most unique scalability problems in the world and work on the newest products we have. It's a mission-critical role that involves a lot of coding and requires a high degree of creativity and troubleshooting expertise, so we're looking for Unix experts and great coders with broad skill sets (like yourself)! We have several positions within the Google.com department that I feel would be a good match for your skills and qualifications. The positions are currently available in: Mountain View, CA Santa Monica, CA New York, NY Kirkland/Seattle, WA Dublin, Ireland Zurich, Switzerland Sydney, Australia

I'm not asking you to quit what you're doing now - just seeing if you might be interested in exploring opportunities with Google. If you're at all interested in the department or in Google in general, I'd love to speak with you about the opportunities we have here - and of course, if any of your talented colleagues would be interested, feel free to forward my name and contact information to them as well.

You also may have noticed that a couple of weeks ago we were named the #1 company to work for by Fortune Magazine: http://money.cnn.com/magazines/fortune/bestcompanies/2007/snapshots/1.html

Thanks for your time, and I hope to hear from you soon!

Best regards,

-{snip}

Top Ten Reasons to Work at Google: http://www.google.com/intl/en/jobs/reasons.html

What It's Like to Be an Engineer at Google: http://www.google.com/support/jobs/bin/static.py?page=about.html

I would LOVE to work for google but even I think I lack credibility.

This is an interesting tip for finding old versions of files you may have knocking about on your hard disk. Assuming you have Google Desktop up and running, if you run a search and select the option to 'See nnn desktop results in a browser' (where nnn is the number of results) you see a typical set of google search results, all the files found, ok that's to be expected and nothing special. However, if you then look at the number of 'cached' entries for each listing you may find that google desktop has kept copies of older versions of your favourite files. If it says '57 cached' then there are 57 older versions and you can click on this number and get a timeline of all the various versions of the file.

Ok, your fancy formatting may be lost but if you just decided that maybe it wasn't a good idea to rewrite your memoirs from the perspective of your cat then the previous version may still be there somewhere.

Someone at work had a nice bird's eye aerial view of the company building, He got it from Microsoft's ripoff of Google Maps which was my route-finding site of choice. The Microsoft version has much better aerial photo's of my town than Google, showing cars where on googles you cannot make out houses.

But it got better: there is a 3D view that somehow manages to show pictures from a choice of the four compass points. The pictures of my house are amazing, you can see the whole estate and compare the size of your garden to everybody else's. I'd like to do some printouts to hang on the wall but am scared of copyright police. The bird's eye shots seem to date from 1999 and my house is shown as woodland. The 3D shots are from last year, 2006, as a couple of houses at the end of the estate are still being built.

What is odd is the consistant lighting level over the whole town. Is it one big picture or do they do some processing on many pictures and stitch them together?

What is it about Barclays Bank customers that makes the phishers target them more than any other bank (according to my gmail spam list)?

• Are they particularly gullible?
• particularly rich and worth swindling?
• is the barclays web site easy to copy?
• are there more than enough profits from barclays that it's not worth copying more sites?
• do the other banks (including mine) hire hit men?

What could it be?

Spam on gmail is definitely getting worse, 30-50 messages a day. I've given up checking the list for false positives beyond a quick scan for subjects that make sense (why would I open an email with the subject 'Re: ceiling banana'?).

I've mentioned before how the current Google Desktop has the nice ctrl-ctrl thing to open the quick search box where you bang in a query term and the results appear in a list. You then select one of the results from the list for further perusal and things freeze for 30 seconds before the results appear in Firefox. The computer is still usable, you just have to wait for the results to appear.

I've found someone else moaning about this in the google forums, it appears to be a Windows 2000/firefox 2.0 problem. No resolution yet but I have found that it is possible to use CTRL-ALT-G to select the google query box on the sidebar and then type your query. This is pretty much the same just not quite as quick to invoke. However, when you choose a result from the list it appears pretty much instantly in firefox, as you would expect.

Another workaround is to use IE as your default browser. The third, more pausible workaround is to buy a Mac and use spotlight.

I've given up on using the google desktop scratch pad and todo widgets to try to synchronise across desktops as it doesn't work at all reliably: better to use email.

So I made some space for more widgets. I install the google video widget out of curiosity and it offers to show me Steve Irwin's death video. Yes, tasteless voyeuristic pleasure always there on your desktop!

I removed it.

The weather globe widget is pretty but my computer room has two windows.

My new monitor has an abundance of screen space that was begging to be used. When apps are maximised they are just too big, there is ample space to have widgets down the side.

I contemplated the widget options:

• Google Desktop Search
• Yahoo Widgets/Konfabulator
• Stardock ObjectBar or DesktopX

I decided it was time to give Google Desktop Search another try. In addition to search it gives me a desktop bar down the right of the screen showing pending email, family photo's etc. It has a number of features that I hadn't tried out before:

• press ctrl button twice for a quick search dialog. Bang in the first three letters or so of a program name and you can launch an obscure program in seconds whichout resorting to the mouse and start menu.
• the widgets include a scratchpad and a todo list. It is possible to store the notes and todo's on google's servers such that they appear at every pc you use. Ok this relies on trusting google so better not put your evil plans in the todo list. UPDATE: it takes a minute or two for notes to appear after booting up.
• Google Calendar widget integrated with google's online calendar.
• they have prettied up the widgets with transparent backgrounds etc. You can drag them off the desk and put them where you like.

I was previously happy with Microsoft's MSN Desktop Search apart from it persistantly failing to stop when I shut down the computer. I'll give google another try but I don't rule out vacillating back.

I've added no real content to the blog for months and let things stagnate. I move servers to save a bit of money and suddenly my page rank goes from 3 to 5. Is it because I am now on a US server and google has a low opinion of the importance of pages based elsewhere?

Do I care since it is probably a glitch? However, if it stays at 5 it could be time to cash in with some ads and affiliate links

I have missed blogging and feel guilty about not updating the site, especially the two pages of comments awaiting approval.

UPDATE: a week later and looking at the sitecounter stats there is no significant difference in traffic.

Today gmail has changed such that when looking at the list of messages I cannot middle-click and open a message in a new firefox tab, I have to open it in the same window. This means I cannot look at multiple messages in multiple tabs.

If I right click on a message there is no option in the context menu to open in a new tab, the only link on the page I have found where this works is the 'help' link.

Hope there is a good technical reason for this and it's not just a stupid dropoff by someone who carelessly breaks features that they don't use. Who are google to redesign web ui standards? Are they going to lower themselves to childish 'disable the context menu because my html is so full of great ideas I don't want anyone to view the source' javascript tricks?

Damn google and damn the free services that I expect perfection from.

I'm having a bout of insomnia and it doesn't make me cheerful and tolerant.

Google personalised home page has got me back into Slashdot. I got bored with slashdot after using an RSS feed that only showed article headlines as teasers to the content. Ironically Google home page only shows headlines.

On a whim I tried logging into my old slashdot account for the first time in what must be four or five years and it remembered me!

I did some metamoderating (moderating the moderators) as it helps while away the time.

Bah, just getting back into using Opera and Google go and release Google Notebook a kind of web based notebook that is best used with a Firefox extension, no such extension being available for Opera yet. The extension allows you to select a load of text on a web page and clip it into a web based notebook. You thus create your own scrapbook of snippets from your surfing. You can add your own annotations, subheadings etc and also edit the text in the snippets. Once you've done this I guess you print it out and you have your homework done.

It could be used from Opera for just taking notes down which is kinda useful but not so much fun and plagiarism would involve bothersome copying and pasting. Too much trouble.

Google Notebook is also similar in utility to EverNote. I wonder how long each notebook page can be?

Update: I know what's missing: tagging. All it has is powerful google searching to help you find stuff.

To get Google Desktop to reindex my outlook email rather than Thunderbird I had to uninstall and reinstall it. Following this it decided to reindex everything every day. After two or three days I got tired of the sound of the hard disk being thrashed all the time and uninstalled it.

Back to no desktop search and not really missing it. Searching email in Outlook is painfully slow and I may well look into another desktop search (msn, yahoo etc) just for this.

I'm back to using Outlook for two reasons:

• the journal feature may be useable as a desktop notetaking tool. Being integrated into the email program means fewer apps running and a leaner and meaner system (ha). It does look like a half decent note taking tool, entries can be formatted, pictures pasted in etc.
• was reading how hackers can carry packet sniffers on memory keys and became paranoid about my unencrypted password floating around the ether(net). I looked into setting up ssl authentication on the exchange server but it involves messing around setting up an ssl certificate server and I couldn't be bothered.

I've given up on using folders to categorise my email: I leave it all in the inbox and rely in searching to find stuff. I have google desktop search to speed this up but that still thinks I am still using thunderbird and tries using that to display messages. How to persuade it I have vaccilated back to the evil ones?

This site wouldn't load. Looking in the status bar it was trying to contact Google Analytics to get a javascript file. This affected Firefox but not IE which loaded fine.

I commented Google Analytics out of the page and now it is working again.

I don't use Google Analytics a lot, I still prefer statcounter for my web statistics because of the detail it gives me. Google Analytics is more about pie charts for marketing people. Statcounter tells me what google searches people are using to find my site and that is much more interesting. Statcounter also gives graphs and averages and these tally pretty much with Google Analytics.

I'll leave just statcounter installed for now, I don't need both, it slows down page loading, google analytics is in the header and brings the whole site down if the server goes, statcounter is at the end of the html and most of the page still displays even if statcounter isn't responding.

Moral: don't link to unreliable servers in your page header.

Was moved to try indexing the company Intranet using Google Desktop Search. I downloaded the kongulo plugin which offers to do this.

It turns out that this is a command-line python program that scrapes a web site for links and submits each one to the google desktop indexing engine.

Well it was broken, kept coming up with the error:

pywintypes.com_error: (-2147352567, 'Exception occurred.',
(0, 'GoogleDesktopSearch.EventFactory.1', 'Component not registered',
None, 0, -2147221502), None)

Going through the developer sdk, this appears to be because the API's have changed and no-one has bothered to update kongulo.

I changed the registration code at the end of kongulo.py as follows to fix this:

   1    try:
   2      # Register with GDS.  This is a one-time operation and will return an
   3      # error if already registered.  We cheat and just catch the error and
   4      # do nothing.
   5  #    obj.RegisterComponent(_GUID,
   6      hr = obj.StartComponentRegistration( _GUID,
   7               ['Title', 'Kongulo', 'Description', 'A simple web spider that '
   8                'lets you keep copies of web sites in your Google Desktop Search '
   9                'index.', 'Icon', '%SystemRoot%\system32\SHELL32.dll,134'])
  10  
  11      oInt = obj.GetRegistrationInterface( "GoogleDesktop.EventRegistration")
  12      hr = oInt.RegisterPlugin( _GUID)
  13  
  14      oInt = obj.GetRegistrationInterface( "GoogleDesktop.IndexingRegistration")
  15      hr = oInt.RegisterIndexingPlugin( _GUID)
  16  
  17      oErr = obj.FinishComponentRegistration() # pcw
  18      # TODO Provide an unregistration mechanism.
  19    except pywintypes.com_error:
  20      # TODO narrow to only the error that GDS returns when component
  21      # already registered
  22      pass
Toggle Line Numbers

I find it odd that Google Desktop Search doesn't natively index intranets (or specified web sites): having to hack command-line python scripts to do it is hardly user friendly. It might be that they want people to buy Google Mini boxes for £2000 a pop rather than hand out free tools.

Maybe they are evil after all?

Incidently, this:

obj.UnregisterComponent( _GUID)

is how to unregister kongulo, as mentioned in the TODO (TODO is a programming term that means 'this needs doing but I can I can only summon the strength to press four keys').

I am back with Google Desktop Search. I was using MSN desktop search but I got tired of it not shutting down with the computer. I have moaned about GDS in the past, particularly that I don't like the idea of results appearing in a web browser because the only thing you can do is open them, you don't have Windows Explorer's right-click context menu. After months of using MSN search I have decided that this is not a big deal: on the whole I can remember where things are and I use salamander to access them, I only use search when I have lost something.

GDS integrates better with thunderbird than MSN search did: I installed an MSN thunderbird plugin but it never worked. GDS supports it out-of-the-box, although as I use IMAP it will only index the messages I actually open, not the 2400 messages in my archives (unless I view them all one at a time).

I have come to appreciate the various plugins available for the GDS sidebar, especially the email preview thing: in fact I find myself reading email on that rather than switching to thunderbird. There are many plugins available but their installation programs often seem excessively large (megabytes) for something that is hosted by another application and appears in a window two inches square.

I have stopped using Opera for now: it's mail reader was not showing me the contents of some messages so I abandoned it. Some emails can crash it and it is still not as compatible with web sites as firefox. It has got me hooked on gestures and I find myself gesturing in FireFox to no avail (except at microsoft.com). I have installed a gesture extension and am hoping the next point release of firefox doesn't break it.

I had an odd comment on this site:

Google PageRank carefully explained and what you can do with it - written by top SEO experts. Here is link for your readers; pagerank-prediction.com/

Going to the site (which had page rank 4) it looked like a legitimate description of the page rank algorithm except that at the end of it were the words "v7ndotcom elursrebmem". Strange.

A search for this phrase reveals a pagerank contest, many sites quoting the above phrase, the one getting top ranking on google being the winner.

To get past my comment filter someone had to submit this by hand. Someone really wants to win that prize if they are going around manually linking everywhere.

Maybe now I've put the phrase here I stand a chance of becoming an also-ran!

google pages appears to give you a free web site that can store 100M of static pages.

See mine here.

• All the templates appear to be fixed width
• No site management: if you want to edit the footer, for example, you'd have to edit each and every page.

Can upload files though.

Here are some sites I am visiting when I am desperate for something interesting and I've exhausted my rss feeds:

http://www.google.com/ig : personalised google home. Add a clock and some flickr photo's, stick it on your second monitor and always know if dell have sent more spam to your gmail account. Add some bookmarks to my other top sites so they are ready wherever you may be.

http://del.icio.us/popular/free : the latest free stuff

http://digg.com/ : interesting articles, although the comments are rather inane. slashdot without depth (no nested comments).

http://tailrank.com/ : this one finds the articles that are most referenced by blogs and hence are presumably the most interesting (or bait the most flames). How did I find it? It crawls this site and I found it in the logs.

http://en.wikipedia.org/wiki/Special:Random : wikipedia random page. May need pressing a few times to avoid obscure star trek characters.

http://www.djangoproject.com/ : see if the magic removal branch has been released yet (django in joke).

Yesterday I did a google for "peter clive wilkinson" (in quotes) and came across a couple of spam blogs that have stolen my content, including the copyright notice at the bottom of the page.

I tried opening the pages but the servers were dead.

Today I did the same search and the old spam blogs have gone but a new one is there.

I won't try to look at the site again as I'm a bit wary of Windows security holes right now and don't want to visit dodgy web sites.

Windows update has installed something or other and is hastling me to reboot my pc. Has it fixed the wmf hole or not? How to tell? I'll install it anyway as I don't like the idea of having gaping security holes on my pc.

Apparently the next generation of processors will have hardware protection against buffer overflow attacks. This might be the only way to protect us from Microsoft's shoddy coding.

Since I started with statcounter very nearly a year ago, it tells me my site has served up 100,214 pages to 72,521 visitors.

I have quickly become bored with Google Analytics: I think it's appeal is more for marketing departments and powerpoint/pie chart enthusiasts. I like the raw detail that statcounter gives me in an easily accessable way. I can see what people are searching for and sometimes it even inspires me to update articles to be more useful.

I still think their counting is buggy: returning visitors seems unreliable, as if page refreshing counts as a return visit, so I take it with a pinch of salt and follow the trends. The numbers are roughly on a par with google analytics.

Statcounter is easier to install, it can go anywhere on the page, Google Analytics has to go in the header. For drupal this means editing the page template rather than sticking it in a block.

I haven't looked at pagerank recently.

Peter's blog finds this site in the first page of google results.

Peters blog finds it on the second page in one article where my punctuation slipped.

Moral: use proper punctuation.

I have set up Google Analytics on this site. This is a very sophisticated looking system for tracking web site visits. Google have set it up to help their advertisers count their money but it is free if your site has less than 5 million page views/month which should suffice for me.

The analytics site is still a bit flaky and overloaded and it is refusing to validate the code on this site. I'll give it till tomorrow to settle down and I'll try again.

It looks to be more biased towards grand overviews than statcounter which is good for analysing individual visits: especially the free version that only tracks the last 100 page views.

Apart from the current flakiness, the analytics reports are only updated every few hours (like awstats I suppose) whereas statcounter is real time.

UPDATE: although the site still tells me the tracker is not installed properly and the link to an example installation is broken, the tracker is working and I can see the stats. So far it looks very sophisticated for a free (to me) service: it looks professional and expensive. Here is the geographical chart:

This shows a tooltip of each town as you pass the mouse over it (the analytics page, not the screenshot above) whereas Statcounter just gives a list of countries. However, in Statcounter I can drill into the visits from each country to see which pages they have been looking at.

I now have a new metric to worry about: depth of visit, i.e. people looking at more than one page

Analytics and Statcounter may complement each other: analytics for charts, Statcounter for the details of who is doing what. I could carry on using both, it's all a matter of how quickly I want pages to load and how much I trust both lots of servers to keep up with mine.

Taking over someone else's pc I decided to install Google Desktop Search again as a new version had just come out. It still has some of the problems I had with it last time, mainly that by showing the results in a web browser you are pretty limited in your options: you can open it, you can open the folder it is in but if you right click you don't get a comprehensive context menu as in windows explorer (or Microsoft's Desktop Search). What I miss here is the choice of editing python scripts or running them.

The google thing does have a new sidebar to display rss feeds and stuff on your desktop all the time but I'm not sure I really need that kind of thing always there at a glance. When my boss buys me a 22" widescreen monitor I'll reserve the right 2" for google.

I installed it on my own work pc but after a while it annoyed me again: if I leave the pc long enough gds seems to gobble up all available memory, pushing everything else out to swap. I then have to wait a couple of minutes for everything to get swapped back in. The other pc I was using had win2k and 512Megs of ram, mine has Windows XP in 256M and swaps a lot anyway.

I tried blinkx as the 'smart folders' feature looked a bit like thunderbird stored searches for files instead of email messages. However, I was disapointed that the folder view again only gave the 'open' option, no context menu. Apart from this the interface was kinda annoying, tiny fonts, didn't index source code, mixed search results with obscure web search results from blogs, couldn't index thunderbird (as gds can), I uninstalled it in less than half an hour.

Maybe the latest Microsoft search thing will shutdown with windows without hanging? If there is an extension for it to handle thunderbird I might go back to that.

I had a play with googles online rss reader. It is very fancy but I don't think it would tempt me away from bloglines:

• You only see one article at a time. Bloglines shows you all the articles from a feed in one page which I think is better for spotting the interesting stuff. You can step through the articles with the j button in both bloglines and the google reader (coincidence? Group homage to vi?)
• I don't want to go through a mix of articles from all feeds in chronological order. I like to go through feed by feed. If I open the list of feeds in the google reader it occupies half the screen on my d410.
• It is a bit buggy: at one point the 'loading' splash thing got stuck on. But like most google software it is beta.
• It carefully strips the colour coding and indentation out of syntax highlighted code. Bloglines strips the indentation. I am not sure why they do this, is it the presentation police?

What I am looking for is the mysthical rss aggregator with bayesian filtering that will learn what interests you and only show that. I'd write one if there were only nine days in every week.

Conclusion: google see rss as the future and want to slap adverts all over my blatherings.

I was studying the statcounter logs for this site and lamenting how the country info only lists the last 100 page views. Because one visitor can look at 50 pages, their country will appear to take 50% of traffic, giving a distorted view of proceedings.

So I decided to update my awstats config with GeoIP to give long term country information. GeoIP is a library from MaxMind that converts IP addresses to countries. Country lookup is free, you can pay for higher resolution lookup (city etc) if you have the $$$.

I installed it as follows:

• Installed the MaxMind GeoIP C Library and perl library from here
• I downloaded and installed the country database.
• I let these configure and install themselves. This may lead to an unholy mix with my debian apt setup but it is more likely to work than if I start messing with it.
• I went into webmin and enabled the GeoIP plugin. A word of recommendation here, webmin does a great job of managing multiple sites with awstats. In fact webmin does a great job of most things.

There is an old version of GeoIP that comes as a debian package but it installs in a debian way and was not picked up by the MaxMind Perl module so I installed the MaxMind stuff by the book.

It seems to be running, it is showing visitors from germany and stuff but I need to let it run for longer and see if the number of visitors from unknown countries goes down (it probably doesn't add country info to old records only new ones).

This may end up simply telling me where the bots that make up a large number of my visitors come from: so far this month this site has used 1.15G of bandwidth. Accursed Inktomi Slurp (which I think is yahoo) has taken 177Mb of this. Googlebot has taken 77M but given me 7073 visitors against 182. Slurp is a good name. It has probably uploaded all the text on the site 100 times. I don't change it that much.

One of my sites is just a placeholder, a default Wordpress installation with a theme installed and the links tweeked. It has been sitting there for a few weeks waiting for me to sort out moblogging technology (damn vodafone scribble all over their mms emails).

I was surprised to see that somehow it has google pagerank 5, two more than this site. Looking at it's backlinks it has 321 (this site has 11 ) and if I go to any of these supposedly referring pages there is no mention of the daily whinge. On the one hand I don't care if the site is accidently pageranked high, the thing that bothers me is that it links to this site and could be giving it a pagerank boost that could go away at any time. Could this be the source of weird fluctuations in google hits? Hits are up by about 100/day since my last vent.

One other thing that I don't understand about backlinks is that dailywhinge is not listed as a backlink to this site although I know for sure that it is.

I suppose google would say it's all beta software...