How Soon Before Gmail Releases Searchable Docs?

The next thing I would like to see in Gmail is searchable documents. Attachments such as word documents, PDF, PS etc would be searchable just like the web search in Google.

A more advanced method that would really rock would work like this. You know the way Google Book Search shows pages of a book after you search for a string...Gmail should treat each document as if it was a book. This would be very easy to do since they already know exactly what text is contained within the document. When you search for something it would bring up a list of emails that contain the string. After clicking on the email instead of showing just the attachment ready to download it should show a link to the page on which this string occurs. Then you can click on the link and it would show the page in a similar format to Google Book Search with the ability to go to the next few pages.

However, with the ability to upload Warez Books to Gmail I think this would be something they would definitely not release since it would just lead to using Gmail as a store for tons of books and material that would just eat up disk space, CPU speed and bandwidth on Google's end.

Collaborative Learning - WikiNotes

One of the wonderful things about wikis is the ability to edit what you want. From an educational perspective I think some sort of wiki should be set up by institutions for each module whereby students can add content to a fixed number of topics.

Instead of the traditional method whereby a lecturer/teacher gives out lecture notes, slides, or some other form of handout they should set up a wiki with a list of topics. As the term progresses and more topics are taught about students should be given certain pages as part of their homework to go and edit. This would give other students a fresh perspective on the subject and at the same time would enable the teacher to see if the student has understood the material.

Once it is published other students would have the ability to go and edit things that they think is inaccurate or pose questions to the class to get more information on a topic. The teacher would have to moderate the whole process to ensure that the correct material is put up and that students have not misunderstood a topic. However, given the nature of most modules this would be easily manageable. Also access would have to be restricted so that only students on the module list can edit...but of course the whole internet audience should be allowed to view the notes.

Why would such a system be needed? Well one of the best ways of learning a topic is to get info from a variety of sources. Allowing a person who is not an expert in a subject but has just understood a topic to explain something to you is often easier to get than some sort of professor talking about that same topic. Also it would form an excellent resource to come back to when you want to revise the subject.

From a search engine perspective it would be an excellent way of getting good easy to access info in an organised hierarchy. At the moment excellent info is hidden away in PDF and PPT lecture notes which are hard to access and often have valuable info hidden away in very concise slides. Getting that information through the social process of wikis would allow students all over the world to learn about topics in a much more digestible manner.

I often use Wikipedia to revise for Computer Science topics since the majority of early adopters of this wiki culture are technical computer guys. Therefore, the quality of info on Wikipedia on subjects like UNIX, Java, Compilers, Models of Computation, Pumping Lemma, Databases, etc are of a very good degree. The way of writing the info is very good to the layman since the info is organised into digestible topics.

However, the problem I have is that since it is written for a very wide audience, often the info is far above the amount that I need. Having a localised wiki which is specific for the course I am doing would be far more ideal since it would only feature things that have been covered so far.

Even for a smaller school this would be a good idea to try and engage the students to do research for themselves and be able to change other people's work. This of course can work for any subject and would work well for team based modules.

Busy!

I've been really busy recently clearing up the backlog of work for uni and generally trying to sort out stuff in my life. I have a few things that I want to blog about...so this is just a heads up that I will be posting again in the next few days. My holidays have now started so I have a bit of time off to explore new programs and ideas. I am thinking about developing a few random programs in J2ME (Java for Mobiles).

Also I need to learn C/C++ since we are starting a new module next semester about UNIX programming, and I know nothing at all about UNIX or LINUX shell commands.

Start Gmail with a page other than Inbox

Currently Gmail does not allow a user to change the folder that is started with. What happens at the moment when you log on is that the first page to be shown to you is the inbox. This may be useful for some people but what if you really want to see your starred folder instead or perhaps a specific label.

Now there is a way to achieve this by linking directly to the page that you want to see. I forward all of my university mail to my Gmail account as it becomes easier to check it in this way. When I am at uni I would like the ability to start Gmail off with my "Uni Mail" label so that I do not have to look at any of the non-uni related emails that I get.

Even better is that if I wish to open each label within a tab then I can also do this with this method. All I would need to do is put all of these in a folder in my bookmarks and using Firefox I would do open the whole folder of links in separate tabs.

So...you must be wondering how this is done. Okay here is the link that you want to bookmark which will give to the starred folder as your initial folder: -

https://mail.google.com/mail/?&ik=&search=starred&view=tl&start=0&init=1&zx=1g852n7uru5m&fs=1

So let us break this down to see exactly what is going on here. The "search" string clearly tells us the starting folder which I have changed from inbox to starred. But what happens if you want to refer to a specific label? The way this can be achieved is to change "starred" to "cat" and then create a new string called "&cat=" followed by the name of your label.

Therefore the link becomes: -

https://mail.google.com/mail/?&ik=&search=cat&cat=insertmylabelhere&view=tl&start=0&init=1&zx=1g852n7uru5m&fs=1

of course you have to change the place I have written "insertmylabelhere" with your specific label. You can include spaces right into the link which will automatically be converted to the "%20" by your browser and will count as a space in your label.

I couldn't actually find a string to do a direct compose, so if anyone knows how to do this then feel free to post the link.

This is a cool method to have multiple views all at once. Especially for those who use their Gmail for "Getting Things Done" and want to check their Gmail account just to reflect on what tasks are left to do and what has been completed.

Google Book Search

I have been very busy with last minute reports and assignments to hand in this week and so have not had the chance to blog. I've found Google Book Search to be very useful for a variety of things as I used it over the weekend.

I've written about Google Book search before on here but I found a few more uses over the weekend when I played around with it. Scanning in images has always been a pain for me especially when you want to quote an image from a textbook. With Google Book Search all I had to do was type in a quote from the page (I had the book in front of me) and the page came up in the search engine. All I needed to do was a print screen and copy the image into my piece of work.

Something that would be cool is if Google provided ready made references for each of the books that they have scanned. A simple javascript that perhaps generates Harvard style or Vancouver style references or provides the LATEX style reference for the text that you have. This would be useful since the user would not have to manually find the right info and then format it into a reference.

Something else I noticed when I used Google Book Search was a flaw in their feedback system. When you click on the feedback button a hidden box shows up asking you to enter what is wrong with the page shown. However, there is no way to submit the feedback. Here is a picture: -


Until that is you scroll down using the mouse...yep the feedback container that shows up is too small to accomadate the whole feedback form and so hides the submit button. Here is the submit button: -


I'm guessing that they designed an original form and then added something in later but forgot to actually change the size of the container form.

Actually one of the most useful things that I found with Google Book Search (whether the content is copyright or not) is the searching facility. When you have a book in front of you it is very difficult to use the index for finding stuff that doesn't deserve a huge section in the book. This is where Google Book search comes in. You search the book using Google and it tells you the page number. Then you follow the page to see if its useful. If it happens to be restricted then you can just open up the book to that exact page and look at it, otherwise you have the page within Google to check if it is exactly what you want. This helps a great deal since indexes in the back of books are so old fashioned and difficult to use for obscure topics.

Oops...after using word so often over the last few days I accidently hit Ctrl+S without noticing...and it published a half written article. Hmm...why is "S" for publishing...surely it would be better to conform to standards and keep "S" for save.

Okay final thing before I go...lots more work to do...Is there any way of searching only for books that have content completely unrestricted? This means that I only want to return results for books out of copyright...I can't see any sort of licencing drop down box on the advanced search unlike the main Google web advanced search. Oh well...I'll have to dig around.

Gmail Mail Fetcher

Yesterday a user mentioned on the Google Blogoscoped forums that Google had added a help page for a Mail Fetcher. I followed the link but it had no body to the help page. It seems that they have now removed the page but it still shows up in the Google cache. Here is a screenshot of the Google Cache version if they remove it: -


A Mail Fetcher could be the answer most professional/corporate people are looking for as an alternative to forwarding. I currently forward all of my university email to my gmail address so that I can get access to it when I am not at uni. If native support for IMAP or incoming POP is included in Gmail you could see a lot of people using Gmail as an alternative to Microsoft Outlook/Mozilla Thunderbird. Sure it would be nice to also have a offline browser based (XUL) Gmail interface but this could be a first step in trying to get a big chunk of the market who only have a pop based email account and have to use Outlook/Thunderbird as their interface.

In my view it would be good for Google to offer a different version of Gmail specialised for corporate use. This would mean that the email would be stored on the company's premises and would be completely searchable with a desktop interface. The web based interface would work in tandum with the desktop version in that they would sync together since the emails would be got directly from the company's server rather than Google's. Of course Gmail would probably need to provide a more secure version (all traffic would need to be encrypted not just the login) to encourage the swap over.

On that subject you can currently get all traffic to be encrypted by using https:// as the protocol. Therefore, instead of going to http://mail.google.com/mail bookmark the following URL to get all your traffic encrypted: -

https://mail.google.com/mail

I think they used to encrypt all the traffic but it caused too big an overhead in traffic to manage on a long term scale since their https protocol goes to a slower server. Now only if you specify the https header then the traffic is encrypted.

Improvements to GMail I would like to see

With the competition concentrating on keeping up with Google its obvious that we are going to see Google try to explore new things with its Gmail service. These are some of the improvements I would like to see: -

• Enable tabs for email -> Having more than one page to read email has always been something that people do especially with browsers like Firefox.
• Allow RSS feeds for specific labels -> So if I have a label called "Uni", which I do for all forwards from my university inbox, I would like to subscribe to that label via the Live Bookmark system on Firefox. At least it would be a better way of knowing when I receive an email from a specific person or about a specific topic.
• Add calendar or to-do list within Gmail -> I think that Gmail should allow a tight integration between email and some sort of calendar or to-do list. Some sort of right click option when you click on a date or highlight a piece of text within an email to allow for adding reminders for yourself.
  
• Change right click to a context menu -> This links in with opening the link in a new tab, trashing a specific email. The menu should reflect exactly what is possible so if I right click on a message it should allow me to open the message in a new tab, in a new window, delete message, archive, apply a specific label, download its attachment(s), etc.
  
• Allow drag and drop into a specific label/folder -> All the selected emails should be able to be dragged into a label. I'm not actually sure how much extra value this would have but as with all good interfaces there should be more than one method to do something.
• Allow direct download of attachments -> First they need to remove the new window that keeps opening when you try to download an attachment. Thereafter from the page that displays all the messages they should enable the paperclip to allow access to direct download of the attachment.
• Show a progress bar with size info when adding an attachment -> The current method (attaches in the background) of attaching the files to a new message is a big improvement on most systems. Even better would be to show the size of the file that is being attached and how much is done.
• One-click preview -> Like one of the greasemonkey scripts written by a Google Employee there should be the ability to see a greater portion of the message just by right clicking on it.
• Mailing List support -> currently there is a hack that I use to achieve Mailing Lists in Gmail. Native support for this would be very cool indeed.
  

These are the UI features I would like to see. In terms of raw features I would love support for IMAP, Google Calendar/To-Do List, and lots of other stuff that I wont go into detail for now. There is a lot of cool stuff going on in the background at Google, I just wish they could concentrate on some of their current services rather than trying to put their fingers in too many pies.

All the official Google Blogs in one place

Philipp Lenssen over at Google Blogoscoped has posted a page and a feed that collects all of the offical Google blogs in one place. So far he has Blogger Buzz, Google Video, Inside AdSense, Inside AdWords, Inside Google Desktop, Google Blog, Google Base Blog, Matt Cutts SEO/Google related posts, Adwords API Blog, Google Reader, Google Code and Google TalkAbout.

It's an excellent idea since people like me who dont want to visit lots of different places can just go to one central location and get all of the news that I want. Note that currently the posts are out of order...but as of now only the latest posts will be appearing and so they will be in the correct order.

It would be good if he was able to generate a custom feed which generates a page depending on what blogs you want to read. I don't mean just the official Google product blogs but also some of the Google employees who blog about a certain product. For example, just say I don't want to read about Google Desktop but instead I want to read Fury.com by Kevin Fox then it should be able to generate a page that contains all of my blogs of preference possibly from his very own compiled list.

In terms of the URL to bookmark, the system could generate a special hash that refers to a specific combination of blogs. That hash would be stored in the database and a script would take the hash and go out and bring the necessary posts and then generate a HTML page. Of course the posts would probably need to be stored in the database too so that they are only parsed once from the RSS and all other times are just got straight from the database.

Gmail to introduce anti-virus

Gmail has added a page to their help page describing how their anti-virus feature works. I don't see it in Gmail at the moment but I think they will start adding this in very soon. At present they block .exe files which to be honest blocks the majority of viruses.

I see Gmail's strategies for viruses /phishing/spam very good indeed. The notifications that are shown at the top for any email suspected of being intended for phishing credit card details or general fraud are accurate and well set out.

The blocking of images has also been a good move in that the whole page does not reload for any email that are not from trusted sources. Recently they did add the ability to permenantly allow images from a specific sender which seems to do the job since most people would only receive emails containing images from a mailing list or from friends.