Blogging from Google Docs and Spreadsheets

Up until now, all my blogs have been done in gmail and then sent to my blog account via e-mail. A further trip is then necessary to check the formatting, fix any broken links and generally check things over.

This posting is different. I am using Google Docs and Spreadsheets to write this and then publish it directly to the blog. Hopefully this will allow me to sort out the formatting and the links in situ, and have the completed post arrive without need of further assistance.

I have used the spreadsheet side of Docs and Spreadsheets before with much success. The collaboration aspect is fantastic for myself and my accountant to work out my tax finances together, on the same doc, at the same time. As changes are made, they are visible to the other person immediately. This makes pointing things out and working through problems remotely as easy as working next to each other.

I can see how companies have taken the Google application suite on-board. Combine the collaboration aspects of this tool with the functionality it offers, and the integration with mail and calendar, and I really can't see a compelling reason to use Office in 80% of scenarios. There will always be a need for complex macros in Excel, but the vast majority of the stuff I do is already available in Google's offering.

Now, to publish and see if this works...

Some command line shortcuts

I know it's been a while, and this is only a short one, but I wanted to get these down.

I've been experimenting with the command line strengths of Linux, combining commands to perform powerful statements. The first is:

find -name "*.JPG" -exec rename 's/\.JPG$/\.jpg/' {} \;

This command will find every file within a tree structure from the point the command is run that has .JPG as it's extension and then rename it to .jpg. Linux automatically recognises .jpg to be a picture but not .JPG. For someone like me with over 3000 pictures in numerous subdirectories this was a huge productivity help. It is simple to modify it for other file types too. Obvious ones spring to mind like .AVI, .BMP, .MPG etc.

Another thing I wanted to do was increase the rights on a tree structure for all files whilst removing execute. I know from painful past experience that chmod -R 0664 * has disastrous results with directories! However, combining find and chmod allowed me to produce these 2 commands:

find -type f -exec chmod 0664 {} \;
find -type d -exec chmod 2775 {} \;

The first only finds files and then changes the permissions to give the user and group read and write, allow everyone read, and removes execute across the board.

The second finds directories and changes the permissions as above (whilst allowing execute because they are directories), and also changes the permissions on the directory to ensure that every new file is given the group ownership of the directory, and not the user creating it.

As I continue to play with regular expressions and the command line I'm sure I'll find more of these gems. I'll probably put them here too.

For now, one more:

find -name "*.zip" -exec unzip -o {} \;

OK, one more:

rename 's/^\d+_//' *

This will remove the leading number and underscore from files or directories such as those that have been downloaded using .nzb files from newsgroups. Useful for tidying up your downloaded directory structure. (Or at least useful for tidying up mine!)

The Google factor

One of my earlier posts (not that there are that many of them!) was on the announcement of Google calendar. All in all i thought it was a good idea. It's been released now for quite sometime, and I've been using it since day one.

It's not the only Google service i use. In fact it would be easier to list the services is use regularly that aren't Google. None that spring to mind (websites (and RSS feeds) not included. I have Google's personalized homepage as my first port of call on the web; i use Gmail as my main e-mail system; Google calendar keeps me from forgetting whats coming up; Google groups gives me an indispensable resource for support; blogger is owned by Google and allows me to spout off to the world; picasa sorts out my photos, and, once the Linux version catches up, will allow me to post my pictures direct to here, and to web albums for my friends and family to view; there is also, obviously, search. All in all, Google has it covered.

So, how do i go about using these services? Well, the personalized homepage is my first stop. This service just gets better and better. With plug-in modules, and tabs, it allows me to see everything that i have an interest in at one shot. On every tab, I have the ubiquitous search engine, and a corner box giving me 1 click access to all my services. My first tab gives me my latest mail, my coming calendar week, an up to date view of my most visited google groups, and some witty quotes. The second tab is all RSS feeds from my most visited sites (and this gets easier by the day as more sites give one click access to their information). I can see at a glance what is going on in local and world news, technology, music, and gadgets / hardware. All with one click access to the underlying stories. The final tab has some jokes, some direct website links (my transportable favourites) and a pacman game for those non-existent dull moments. All in all, a veritable one stop shop for the web.

Gmail started off great and just got better and better! I have my e-mail where ever i have Internet access. It is all categorized automatically, extremely accurate at detecting spam, and searchable in any way i have needed to date. I have just downloaded the Gmail app (you need to use that link from a mobile phone) for my mobile phone, so i can access my mail anywhere i have a signal. (This and the last entry were both written on my phone). Gmail also gives me direct access to Google chat. Not my most used feature (most people i know are on MSN), but still useful when clarification is needed on a recent e-mail, or a conversation needs more back and forth than really suits e-mail.

A number of years ago, before Google was synonymous with search (yes, such a time existed!), there was a web site called dejanews. This was a great resource for accessing newsgroups from a website. An absolute necessity for office support workers. Newsgroups are an extremely valuable source of real world information. Google bought this company, and slowly re-branded it Google groups. In time they added mailing lists and posting, expanding this already fantastic resource. Google groups is always my second port of call for support once a normal web search has failed to turn up an answer.

As I was hoping in a much earlier post, Google calendar has been the answer to the majority of my needs. I have 2 calendars, and access to my wife's. Between these we can keep our lives in check. As already mentioned, Google calendar plugs into my homepage. There is also text messages whenever an event is due to start. Wherever I am I know where I need to be next!

I am also signed up, obviously, to blogger. With the new release this is tied to my Google account, so I have single sign-on to my services. I also have access to docs and spreadsheets, a portable version of basic office functionality. It's not there yet, but functionality is increasing all the time. Charts in the spreadsheet app will make a huge difference, but the lack of code will make it impossible for me to use frequently.

On the sidelines is Google earth and picasa. Google earth is pretty, but not something I use much. I am glad they ported it to Linux though, eye candy is always welcome when showing off an OS that isn't windows! Picasa, on the other hand, is an app coming into it's own. With the addition of picasa web albums, and the direct access from one to the other (hopefully coming soon on Linux) you can keep all your photos catalogued, and share them with your friends and family from one app.

I've touched on Google chat from an internal mail perspective. I also have Google talk and Google desktop installed on the majority of windows boxes i use. These 2 tie together to let me search all my local documents, the web, my local mail, and my gmail from one easy interface. They also inform me of new mails, and new news on my frequently visited websites. Not as heavily used as it could be ( a FireFox tab is always available for searches), but part of the whole package.

Finally, for this post, there is search. What more can be written on this?! I can, and regularly do, search the web, individual sites, pages, images, news, purchases, newsgroups, mailing lists, videos, and probably a couple I've forgotten. One worthy of a mention is Google local. Allowing me to search for businesses in an area easily, and tying it into a detailed map of the area allows me to plan excursions and awkward shopping trips far easier than I used to. I know that without Google we'd all still be using Yahoo, or Altavista, or Hotbot, but would it be as good? Probably not.

So there you have it. My life is run by Google! Do I feel happy about this? Yes. Would I let anyone else do it? Probably not. I for one hope they go from strength to strength, and the functionality just keeps getting better.

Downloading files from usenet

One of the most common questions i get asked is how to get files off of usenet and into a useable form. These files can be any number of things, but the purpose of this entry you can think of them as legit. :)

A little background first. Usenet predates the web, and allows anyone to post information to one of many newsgroups. These are distributed all over the net, and are synchronized automatically. Everyone has access to usenet via google groups. It is easiest to think of them as large forums or mailing lists. All usenet groups have size limits on the files posted up to them. For text groups this isn't a problem, but before the web and downloading there was no easy way to transfer files to other people (this was well before people ran there own servers or had "always on connections"), so someone came up with the idea of converting a file into a number of smaller files, converting those (in a reproducable manner) into text and posting them onto the newsgroup. Anyone interested could then get all these bits of file, re-combine them, and turn them back into the original file. This was at a time where if you were connected you knew how to do this. Geek would be the word you're thinking of!

Fast forward to now, and usenet still exists. It deals with huge amounts of traffic every day, and it is still a repositry for files. The difference is that normal people now use the internet, and they haven't a clue what usenet is, never mind how to use it! That's where this comes in.
There are a number of ways and means of doing this, but i'll cover the ones that work for me. First, you have to find the files. Each newsgroup is made up of millions of messages, and each message has a header. Think of this header as the to, from, and subject part of an e-mail. If you want to know what's in a newsgroup you have to get these headers. As you may need to go into a number of newsgroups, and each one has millions of headers, you don't want to do this. You want an index. Searchable, simple ways of getting exactly what you want. Luckily one exists. Go to newzbin. Here comes shock one. You have to pay (it won't be the first time) if you want to make this useful. You can search newzbin within any of the categories, but that will only get you the existence of the file and the newsgroup it's in. If you're cheap, you can go to the newsgroup and get all the headers, then select the ones you want, and download them, but that is a lot of hassle. It is better to pay newzbin (about 20p a day if i remember correctly) to become a member, you can then get an option to download a newzbin file (extension .nzb) which contains exactly what you need to get the file.

Once you have this file you need to use it, for that you need a program that can read and understand it. I recommend one called binary newsreaper. Once this is installed you need newsgroup access. You probably have this through your internet provider, but that probably doesn't have binary newsgroups and will have terrible retention (how long you can get a file for, newsgroups are purged after a period of time) and completion (there is no way to know if something is there until you try to download it. Bad completion means lots of wasted effort). Once again, it's time to part with some cash. I use newshosting which offers unlimited downloading for $14.95 a month. Pay up, get the access details, and follow the newsreaper instructions to get connected.

Once you're up and running, newsreaper has an "import nzb" option. Choose it, select your nzb file, a destination, click ok and then watch it go. It may take a while depending on the size of the target file.

When it's finished downloading you can look in your destination folder. You probably have several files ending in rar, par2, or r and a number (r01, r02 etc). Not very helpful. Step one is to confirm all the files are complete. That is what all the par2 files are for. Download a free program called quickpar, install, and you should then be able to double click a par2 file and have it automatically check (and repair if necessary) all your files. Once that has completed, get a copy of winrar and the rar files can be clicked. Always choose the .rar file, or the part01.rar file. The program can deal with it if you pick another file, but it's best to make it simple.
The unrar process will leave you with you file(s). All the par2, rar etc. files can be deleted now.

(K)ubuntu update

It's been a lot longer since my last post than I had anticipated, but a new job, and a child on the way conspire against my free time!

I said I would update once I had established myself with kubuntu, so here I am. The experience has been interesting. Initial installation was a dream, one click, 6 questions, easy enough partitioning (managed to keep my home and data areas whilst wiping everything else), and away it goes. First impressions on first boot, it's not as pretty us SuSE. The boot loader for kubuntu (or ubuntu for that matter) is a basic text with DOS style letters. SuSE has a nicer box and a more graphical feel. Once booting, SuSE gives you a nice splash screen, and gives you the option of pressing escape for more information (which appears bordered by the SuSE splash screen headers and is colour coded, uses a nice font and is exceptionally well laid out). Kubuntu gives you a splash screen that looks like it was knocked up in 30 seconds, some default information as things start (but no information as to how they got on), and then half way through the boot, the splash screen disappears and a text screen appears using a 640x480 resolution which can't fit the information on to the screen,. The information is just the default linux messages, but with no formatting, no sizing. Basically, it's a mess.

Once booted and logged in (default KDMs are as nice as each other and irrelevant as I change it to my choice anyway), I was presented with my desktop. Thankfully this was the same one I had under SuSE, so the settings had transferred nicely. A few things were missing, but as I had added them to SuSE I wasn't surprised with the need to add them to kubuntu. And there in lies the only major difference between the systems.

Package management (and when all else fails, manual package installation, or, heaven forbid, source compilation). SuSE uses RPM, kubuntu uses DEB. This is irrelevant. kubuntu uses adept, SuSE uses YAST. This is the difference. There is much debate on the web over the brilliance that is the .deb package management system. It is, without a doubt, superior to .rpm, but you don't need to know this with default system packages. Everything should have been chosen and tested to work together, so all that counts is the ease of choosing applications. The first thing to realise is that non of the distributions come with packages to do everything you want. The most glaring deficiency is the lack of support for most media, meaning you have to add new repositries. Within SuSE this is fairly graphical and easy enough. Within kubuntu you are editing a text file. It may try to pretty things up, but essentially you are editing sources.list. I don't mind this (in fact I edit the file manually at the command line), but it isn't very user friendly.

Side note: I have now found an application called easyubuntu which modifies your sources for you and installs the applications you probably want. It's still not as easy as it could be.

With your sources modified you have to choose what you want to install. With both distributions, there really is a need to know what you want. To me though, SuSE's search options were more intuitive the kubuntu's. Once selected, things install and generally work.

What happens if you can't find an app in the repositry? You need to install it manually. Within SuSE and kubuntu if you find an rpm or a .deb, then things are easy. Double click the file and away you go (not strictly accurate, kubuntu defaults to unpacking the file in ark instead of installing it, but it is negotiable). If you only have source to go on here comes the fun part! Within YAST there is a helpful option on the menu that allows you to install all development header files for all the libraries you have installed. One option, one click, that's it. Kubuntu needs you to know what they are all called, and then you select them manually. When you are compiling something complicated (for example hugin an open source panoramic photo editor) there are dozens of library files needed, and no one tells you up front what you need. You have to run ./configure again and again letting it fail slightly further each time as you install the dependancy it's missing. Boy is that fun!

And that is it. For my own use, the distributions are as similar as they need to be. Some things are in different places, but it's easy enough to work out. Both distributions are capable of running everything I need. There is slightly more support for (k)ubuntu the SuSE, but not enough to worry about.

There is a 'but'. Isn't there always! I have an AMD 64 bit chip as my main machine's processor. On SuSE and kubuntu there is equal support for the main OS. However, application providers don't have the same support. Within SuSE they get round this by installing the 32 bit libraries alongside the 64 bit ones. If an app can only run on 32 bit, there isn't a problem. Kubuntu doesn't have this option. I ended up wiping the OS and installing the 32 bit version. It hasn't made a difference to performance, but it doesn't feel right. It 'should' be easy enough (I don't know I'm not a developer) to install the 32 bit libraries, why can't kubuntu do it. I did find instructions to setup a '32 bit chroot', and tried this, but it broke after an update, and it was too much hassle to keep up to date.

For all it's failings, kubuntu does seem to be a good OS. It fixed the problems I was having with SuSE (but so have SuSE apparently), and I won't be switching again soon. Stay tuned to know how updating to the next version of kubuntu goes later this month.

SUSE 10.1 and a journey to (k)Ubuntu

I write this with a bit of a heavy heart (not sure why, it's only software!).

I have been an advocate of Novell software my entire working life. I worked with Netware from 3.12 right up to 6.5. I have setup remote access with iChain, installed eDirectory into Windows environments, and, most recently, when I decided to use Linux as my main OS, I chose SUSE 9.2 as they had just been bought by Novell.

I used 9.2 successfully as a dual boot system with XP, and after a successful upgrade to 9.3 I finally put XP onto a VM within Linux and changed to use 100% Linux as my OS. I spent many a happy hour configuring 9.3 and KDE to be exactly the way I wanted it to be, and had the system setup to play music to fit the mood, surf the web the way I wanted, seamlessly switch between all my systems, access my remote file systems as part of the OS. In short, the system worked, and worked well!

I awaited the release of 10.0 with great expectations. I had recently discovered the fun of additional repositories and had updated my system to use the latest toys, but the underlying system was causing me a few problems (USB, networking) and I hoped these would be fixed in 10.0. On the day of release I downloaded the system CD, booted selected network install and upgrade and let it go. I hit the problem that has plagued SUSE (and, from reading around, all distributions including the great Debian which tests and tests until it's right), upgrading doesn't work! Things get left behind, dependencies are problematic. It's not just with the stuff you've installed from source (Although that really doesn't help), even things that have been installed as rpm via yast don't work. Problems really become apparent when you have something newer on your system than is in the distro you are upgrading to. Long story short, my /home is on a separate partition (by design) and a could do a new install without blowing that partition away, but formatting everything else.

10.0 was great! Not perfect, but pretty close. Over the months, I upgraded it to KDE 3.5.2, got the latest amarok, sorted out playing every media format that the web threw at me. In fact, the only problem remaining (which can't be fixed) was flash on the Internet. Maybe, one day, macromedia will finally release a decent flash player for Linux. Until then 7 is what we get, and therefore some things don't work.

As 10.1 approached I was interested in the changed package system. Yast is good, but a bit clunky, other updates were also enough to get me to upgrade again. Surprise surprise, same problem on upgrading as last time! It was expected this time, so one clean install later and here we are.

My experience of 10.1 is not all great! Some things work well. Dependencies with openGL were fixed, changes in the kernel fixed a few niggling hardware problems. Updates to core components fixed some problems I had been having. Updating on the other hand died a death.

As far as I can tell. Novell wanted their Zen suite installed into the OS, but left it late in the day. Zen has been around for years, I've used it from version 2 onwards in one form or another. It's fairly good as a package management system and desktop management tool. What they have put in to SUSE 10.1 is not the Zen I know! There is a botched merger of Yast and Zen that in theory allows you to use the old Yast tools, and the new Zen package system to configure your updates. At first it seemed quite good. A guide I found on configuring it allowed me to add my main repositories and the updates for KDE and Gnome, the system informed me of the updates, and installed them. Rebooting brought them into play, and also removed the update icon. I can't find it now, it's gone! If I head into Yast I get numerous messages about unsigned updates (I'm sure that will change as the repositories get signed and certified), updates are then selected as they used to be. However, if something goes wrong (a package isn't in a repository that reported it had it (something that happens far too often)) Yast dies. Immediately. If you use 'rug in' from the command line, then things fair a little better, and installs work. So far, 'rug up' reports there are no updates, but I live in hope. Every so often though, a rug command results in 'Waking up ZMD...' and that's it. Good night Vienna.

The other show stopper for me is the install of ffmpeg on the packman repositories (although the bug is also in the latest CVS). I have a script that generates DVDs for me, and uses 'ffmpeg -t 5' (and some other switches) to create an empty audio track for the menu. -t is now ignored and defaults to 10000000000 seconds! Not much use!

The main problem is the lack of interest and information from Novell / SUSE. The update problem is a HUGE issue, and it doesn't work anything like as described. An update system that doesn't update is no good to me. Never mind installing new packages, what about security updates. Will it notice them, will it install them. I have no idea, and neither does anyone else. This is not what I want from an OS, so where do I go?

Ubuntu is the distro of the moment, huge amounts of publicity, huge amounts of investment. It's Debian based. My server runs Debian stable. I like it, it works! The update mechanism is one of the most tried and tested systems around. The latest version ( 6.06) has been released, and I'm going to try it. It is available on a 'Live' CD which I'm downloading now, and there are well established updates to take it to kubuntu so I can keep my KDE desktop.

Expect an update soon!

Remote access to your desktop

By far the most frequent question I get asked is, "How can I access my desktop remotely?". I usually, quickly, remind whoever is asking that they mean, "How can I access my desktop remotely and securely?"! And there you have the crux of the problem. What is secure? How do you achieve it? How can you set something up that allows secure access without making it so complicated that you never bother using it? Here is my solution. Unfortunately, it is not a "next, next next, finish" type solution, but it is as close as I can make it.

I always break the question up into 2 parts, secure access, and desktop access. Once you have achieved one, the other is easier. The answer to part 1 is SSH. SSH stands for Secure SHell, and gives you exactly that. For those that don't know, a shell is a command line interface without any of the graphical functionality, but with all the power of the system (not quite if you are running windows). By installing and setting up SSH you gain login access to the system.

I was going to write this for Linux and Windows, but have concentrated solely on Windows for the time being as there are differences in the setup and configuration of Linux (not major, just different), and the majority of people who ask me about this are Windows users.

SSH installation instructions for Windows

I first set this up when I was using Windows as my main operating system. When I first set this up, it involved a very convoluted process with cygwin and Linux binaries. As things moved on however, it got easier, and now all you have to do is go to sourceforge and download the latest binaries installer release (version 3.8p1-1 20040709 at time of writing). Once downloaded, run the installer, accept all the defaults, after all the files have copied, you will get a message about editing the passwd file. Click OK to this and then finish. SSH is now installed.

SSH configuration instructions

The next step is to setup a user account that will be used for logging in to the system. I can guarantee that once the system is open to the Internet you will get hundreds (if not thousands) of login attempts every day from automated script routines running against long user lists. By using a user ID that doesn't sit in one of those lists, and only allowing that ID, all of the attempts against your system will be in vain. The trick to picking such an ID is to use a non standard name or word. Try putting your first and last name together (eg stevenixon) or putting a couple of numbers in front of your surname (eg 23nixon) it doesn't really matter what you go for as you won't have to type it in, as long as it isn't guessable. Obviously, the ID should be given a password! That password should be as secure as you can make it. There are a few schools of thought on password strength. Essentially, it comes down to what's easy and yet secure. Putting in a password like 'p455word#' is relatively secure. It's longer than 7 characters, and it uses a mixture of letters numbers and non-alphanumeric characters. A password like 'This is my password.' Is also good. Which ever you use is up to you, whatever you find easier to remember and type. Just remember, you won't be able to see the password as you type it.

You now need to create a passwd file in c:\Program Files\OpenSSH\etc\ with the correct user ID entered into it. To do this, go to a command prompt, change into the c:\Program Files\OpenSSH\bin\ directory and run the following:

mkpasswd -l -u UserID >> ..\etc\passwd

Substitute the user ID you have setup for UserID.

Firewall settings

Most people have a firewall these days. Whether it is a NAT router, or some software on the machine. In all cases, port 22 must be allowed to reach the machine you want to control. For NAT, a rule must be created to forward the port to the correct IP address, for a firewall on the machine, the port must be allowed to talk to the sshd service. For instructions on doing this, please refer to the documentation that came with your router or firewall software.

Gaining access

The vast majority of people who want access to their machines are coming from Windows machines. The easiest and best solution to remote access for what we have setup is a free program called putty. Download the latest version ( 0.58 at time of writing) and unzip the contents to your USB drive into a folder named putty.

From a machine that isn't the machine you want to control, run putty.exe. Ensure that SSH is highlighted for the protocol and then enter the IP address of the machine you want to access (If local then this will be the internal private address on the machine (eg 192.168.1.2). If over the Internet, then this should be the front facing IP address that you get from your ISP). Then, under Data which itself is under Connection, put the User ID into the Auto-Login Username box. Under SSH put a tick in Enable Compression, and click 2 only for the preferred protocol version. Now click connect.

If all goes to plan, you should get a dos like box with a security banner printed on it, and a prompt for your password. After entering your password you should be at a dos prompt, in your home directory. Type 'dir' and you should get a directory listing. Job done! You have securely accessed your computer.

Remote control

As promised, once you have gained access to the computer, getting remote desktop control access is far easier. Download a program called vnc (at time of writing, 4.1 was the most up to date version). Install this onto the PC you want control of. Setup your firewall to deny incoming requests to the vnc service. Finally, put the vnc client software onto your USB stick with putty.

Now, when setting up your connection to the machine using putty, navigate to Connection - SSH - Tunnels, and enter the port number 5900 under source port, and 'localhost:5900' in destination port. Then click add. When you connect and login now you can run the vnc client software from the remote machine and put in 'localhost:5900' as the destination address. When you click connect, you will be remote controlling your home PC.

If all this worked, congratulations! This has taken far longer to write than I hoped, and as such has been done over a few days. I apologise if it is incorrect, if you let me know, I'll update it accordingly.

Google Calendar

The web is awash with news that Google is apparently working on its own calendar (CL2) program to integrate with Gmail. The closed beta is ongoing with about 200 participants of which I am unfortunately not one. I am, however, very excited and interested by this news.

I have been looking, unsuccessfully, for some time for a collaborative calendar application that I can access from anywhere and sync with the many pieces of technology I carry with me. I must admit I was probably approaching the problem in the wrong way. I carry a mobile phone with me at all times (who doesn't), and it has a simple calendar function built in. In theory this synchronizes with Microsoft Outlook (a product I don't use unless I have to). I also have a Compaq Ipaq (which now mainly runs Tom Tom software), several computers running various operating systems (the main one running Linux), and a wife-to-be who needs to be able to access and input into the same calendar.

I tried synchronizing all these solutions. If you feel tempted to try the same thing, one word, don't! There are replacements for Microsoft's Activesync available for Linux, but they have varying degrees of success. Even staying proprietary and keeping everything Microsoft, the entries entered directly into the phone would be ignored or even wiped out by synchronizing with Outlook as would the entries in the Ipaq (which also wiped the phone if done that way). As for the Linux computer, forget it!

I then tried collaborative web solutions. Tiki Wiki seemed the way forward for the 3 days I wrestled with it. However, if I found it overly complicated, the my better half wouldn't have a hope!

I have long been a supported of Google. Their search is by far the best out there. I have transferred the majority of my mail needs to their Gmail application. I use Google desktop when I am on Windows machines (I wish they would produce a version for Linux). Picasa is a fantastic application for dealing with photos. The Google gateway software is my home page wherever I am. The ability to aggregate all my normal content into one page where I can get uptodate from one screen is fantastic. The accumulation of text based news-groups into an easily searchable format is by far one of the most used Google applications for me (even if they encourage top posting and not including the previous post by default)

So, when I heard that Google was doing a calendar, I was excited. Not only a calendar, but one tied into my mail application of choice. With luck, this is the application I have been waiting for. Something I can use from anywhere, can allow other people to view and access, and cross check everything with the e-mail or conversation it relates to.

So what does it look like? The main screen is below


The create event detail can be viewed in more detail here:

You can create events on your calendar in a number of ways. (The following is from the CL2 trusted Tester Guidelines:

Click 'Create Event'

• This brings you to the create event page, where you can enter information about your event.

Click on Quick Add (or type the letter Q)

• Quick Add gives you a text box where you can type all the information about your event in normal English, and we'’ll fill out the form for you. We'’re pretty excited about this feature, so please let us know how it works for you.

Drag-to-create

• If you'’re looking at the day where you want to create an event on your calendar, just click and drag your mouse from the desired start time to the end time. Once you've selected your time range, you can just choose a title for your new event.

Event Pages

• Whenever you create an event, we create a web page which you'’ll see when you click on the '“more details' link on any event. This web page is only visible to you, unless you'’ve invited other people to your event or made the event public, in which case you can use the page to share information about the event with people who are attending or the public at large. Note: you don'’t have to be a CL2 user to be able to see event pages, so you can use these pages to share information with anyone involved with the event, regardless of what online tools (if any) they use. (Ever wish your favorite local band would learn how to use HTML and publish their calendar? Once we get your feedback and open CL2 to the world you can help them do just that.)

The following are various setting pages for the application:







That's all the information for the moment. I for one will be watching this very carefully, and as more news is forthcoming, I'm sure to comment on it!

Windows Live search

Windows Live search went 'live' today. Now, OK, it's only a beta, but it is released to the public, it's on display, and easily accessible. It's also accompanied by words from Microsoft that are give the idea that google should be quaking.

So off I trot, and what do I find? So much JavaScript and extra rubbish the page takes 2 minutes to load! Remember this is a search engine, and is directly in competition with google. When was the last time Google's homepage didn't open quickly for you?

I sat there looking at 'Loading' (with a lovely spinning logo) for a good minute and a half before the search page finally loaded something that looks like a bad rip of Google's gateway idea. ( www.google.com/ig for those that haven't been there is a great way to collect RSS feeds together with your gmail account, and the links you use frequently, and present them to you in an easily accessible manner).


Finally I get to search for something. Now what should I search for? It's a Microsoft search engine, so the obvious things were Linux and Google. What did I get? Well, more 'Loading' initially! Then some results. First thing that strikes me is the locality of the results. (well, actually it was the slow speed of the page, but I think you're getting the idea about the speed of the service!) The first 2 or 3 results are all .uk pages. Whilst this is admirable, no one is going to say that the most authoritative pages on Linux or Google are in the UK! If I was searching for an item I wanted to buy, then nationalized results are what I want, but not in every case. Direct comparison is with google of course. They re-direct me to google.co.uk on load, but give me the option of 'pages from the UK', and default me to 'the web'. Exactly what I want.

Second thing that strikes me is the way of displaying the results. There are 8 results a page, but those pages aren't real, I have a weird JavaScript scroll bar next to them, clicking (or dragging) this brings the next 8 results up further click brings more. As I flick through I get the third thing (and possibly the most annoying). The ads at the side of the search results (looking suspiciously like Google's!) change as I scroll, but unlike the results they don't do this is a 'nice' visual transfer way, they just change. I don't know why this annoys me, but it does.

So, down to results themselves. I've already mentioned the locality, so for the majority of my searches google is more accurate for the first hit. Google also provides more results in the main, but I find that pointless as a statistic. Who cares if one returns 800000 and one 10000000 I'm not going to search through more than 20 before I re-try my search, and I already know which is quicker to manage that!

Restricting searches with site: works as expected, but other google tricks like definition and maths problems don't. Currency conversion isn't there either. Obviously there is no local results at the top of the page for businesses. There is obviously no image, news, group (newsgroup) search. In short, it's just not there as a competitor.

So, Windows Live? Windows Terminally Ill more like! I don't think google will be quaking any time soon. And I for one won't be moving away from google!

Brief Outline

So, yet another IT geek spouting off on the web. Well, why not. But why? Read on....

I've been reading tech websites for years now. I have RSS feeds automatically collecting all the stories I have for me, I have more sources and opinions than I can possibly read. I belong to some of the webs most used web communities, but what I don't have is a way to be heard! Yes, this is all about ego ;)

I am probably more aware of technology than anyone else I know, and I have many opinions. However, posting on a site like Slashdot is a pointless exercise. By the time I get there, there are 20 odd top level posts, and everyone is arguing the finer points, the moderating kicks in, and anything added at this level now just disappears. Of course you are supposed to add something to a post that's there, but what's the point. I often don't want to agree or disagree with any of the top level posts, so no point posting there, I want to make a point of my own.

This all leads us here! I don't suppose anyone will read this, but it makes me feel better. New techs that come along, and interest me, will get commented on. The stuff spouted is my opinion. Maybe someone will find it interesting....