Utills Thoughts and Ideas

Saturday, February 12, 2005

Sign-up to search engines

During the past week I have been setting up my controversial blog site. To bring traffic to it, I signed up with google® and yahoo® but have so far failed even though this blog is now listed on the search engines due to my site. While I was looking for the links to free url submission pages of search engines like google® and yahoo® I came across the idea of writing a plugin for firefox which allows the user to submit their urls to these search engines with a click of a mouse.

That would make everybody's life so easy because person creating the plugin can setup an xml backend which will have the links to all the search engines that allow url submission and the url is sent to all those that the user wants to send to.

Hope you guys understand my blabbing because I am really tired and its been a long day at university. To make it worse it is now 12:25 am and I don't feel good.

Thursday, February 10, 2005

Firefox reaching cult status?

The recent blog entry on Asa's weblog got me thinking... The entry itself is about auctioning off a copy of Wired magazine (the one with Blake Ross on it) which is autographed by some of the lead engineers. Now the question that came into my mind is that, is firefox becoming one of those products which have cult status? If so...is that good or bad?

The big problem in my view with this is the fact that the product/company becomes revered in the eyes of a few who then see the product not in with a critical view but instead see it from a biased perspective. If firefox wants to become accepted to the common person it has to lose this public image of an open source product and become ready to be criticised in the way that purchased software is subjected to. However, in my view this is exactly where Firefox has excelled compared to other open source products. The good balance has been achieved by the technical staff who make the stuff, and the marketing machine out there who promote its benefits (not to mention most open source products are backed with huge budgets - or interests from the likes of IBM / Google / AOL). Too many products in the open source arena never hear their name outside small circles mainly because they are too difficult to use by the average joe bloggs. I mean who really has the time nowadays (aside from computing people) to try out a new product if it requires time and effort to understand? We are a society of "if it is easy i'll do it" and so unless it reacts to common human metaphors we are unlikely to use the product.

When looking at some of the products who are seen in this way I see Apple and Linux very much as perfect examples. Taking Linux, for people who understand and use it...they could talk for hours about its FUNCTIONAL advantages over windows, but at the end of the day.... for a person who just wants to use the computer to surf / watch movies / listen to music ...Linux ultimately fails in its HCI (Human Computer Interaction). I can just see all the linux lovers screaming that Linux has built in video player / music player / open office / browser etc but at the end of the day are they as intuitively designed as say Microsoft?

Sure, I agree that MS lacks in the security department but no one in their right mind (apart from mac users) can argue with the fact that the usability of MS products or more appropriately the ease of use is the reason for their success. Its not usually MS who invent all these new technologies and innovate....no they take a decent idea slap a more than decent interface onto it and then market it like hell.

If open source technology started doing that...i.e separating functionality from GUI then I believe we'll see a much greater uptake of such products.

As for the cult status.... the relevance of the above two paragraphs is that MS only innovate if they see a competitor going for their throat.... If some product has cult status they are unlikely to react as they wont see it harming their mass audience.

Firefox in my view is getting out of that bracket and is becoming much more common place as it is easy to use and functionally superior to MS IE. However, until they do not do more to help push it towards adapting to the commercial sector, then i cannot see a great uptake in that field. The argument for ActiveX in corporate environments cannot be solved by saying... just install the ActiveX extension some guy has made... why? well because companies want tried and tested solutions. They dont want to commit to firefox only to find out that some part of their intranet is now disfunctional because it used IE specific tags or that it had badly coded HTML which IE accepted but Firefox didn't. Compromises need to be made....but the simple solution is to package up the extensions relevant to different users and then offer them as being integrated into the download of firefox. Its a bit like selective install but do it on the site so only the selected extensions are bundled with the download.


Tuesday, February 08, 2005

Intro:

Ok, ermm. I am a friend (o_0) and co-conspirator of Utills. I will blog here sometimes when I feel like it ;) or when Utills forces me to.

So, as this is my first blog, I would just like to say that I am a Java, HTML, ASP, JavaScript, VBScript, Visual Basic pro. Just to top it all off, I must add that I am a MCP as I have passed my SQL Server 2000 Installation, Configuration and Administration exam.

Gotta run. Need to do some teamjava work...

Monday, February 07, 2005

In response to possible scam

In response to the article by Yahoo, regarding unicode characters which mimic genuine site names the obvious solution is to use a highlight system on all abnormal symbols.

e.g. if the browser was to see anything outside the range of ASCII codes....then just hightlight the chars....or bring up an alert....asking the user to check the certificate!

ọ could be highlighted to show its not same as o. i.e www.oooọooo.com becomes www.oooooo.com

" A fix won't be easy because the vulnerability, publicized at a weekend hacker conference, that enables so-called "phishing" scams involves a feature, not a coding error."

All of these phishing scams involve some aspect of tricking the human by virtue of the flaws with human sight / lack of precision.... so the obvious solution is give the tools to the end user to decide....dont let the code run in the background blocking all these sites..... furthermore, as most people are unlikely to be using more than 3-4 sites where they deem security to be of utmost importance...(a few bank accounts / email accounts / other stuff) then y not just add a feature to make these trusted....

Updated :

oops slight misunderstanding i think....i thought they would allow u to have any char.... instead i think now that they will have same chars....but cos the same char e.g "a" has different codes in Unicode and other maps like Cyrillic then the computer would know...but the human would see no difference...

" Engineers have rallied around a character system called Unicode. The newly discovered exploit takes advantage of the fact that characters that look alike can have two separate codes in Unicode and thus appear to the computer as different. For example, Unicode for "a" is 97 under the Latin alphabet, but 1072 in Cyrillic." (I would like to add that is 61 and 430 in Hex respectively....try it in Windows char map...under advanced)

But still my argument regarding highlighting holds up....cos it shows the user that they are different!

Quick Update Blog

I've been looking mostly at XUL stuff this weekend... and although i haven't done much work, being the lazy git i am, it all looks pretty damn easy.

All XUL (pronounced Zool btw) is, is an xml doc where a tag like

button id="unique-name" label="text in me"

creates a new button....

(N.B. i have left out the "<" and "/>" tags out cos otherwise it makes a real button...cos i'm using firefox methinks)

so with that sort of ease of use it looks like the redesign of the bookmarks in firefox is back up on the agenda...

am still struggling with Team Java stuff though....dunno what we are heading towards....but i have made a simple tree walker that basically just goes to each node and does nothing! On the plus side i did have a good idea about reforming a line from the tree using column numbers, but seems like there is an easier way if we can get a line number....cos u can just access the file directly and pull out the whole line! dont ask me about effeciency though...cos both seem pretty expensive unless i remake that tree data structure.

As for the update....I will post some screens soon once i get XUL Maker up and running....otherwise it'll mean coding all by hand!


As for topical news....Gmail has started giving out lot and lots of invites....i got 50 invites this weekend....so if anyone wants one then drop me an email at utills+gmailInvite@gmail.com.

Just a thought struck me now.... imagine i signed up to all 50 invites as my own... and then used the Gmail Storage Tool to use it to store files... now there's a free 50gb worth of storage!!

Wednesday, February 02, 2005

Pet Project Ideas

As I mentioned in the previous post, I will soon start on a pet project to keep me busy over the next few months. Although I do have some current uni stuff which may develop into an interesting tool, I plan to keep that separate from my own path into the development world.

Just for interests sake, if you are wondering what this tool I am doing at uni is, its something that would help a programmer understand code by use of structured presentation and / or diagrams. If you are interested then post a comment with some ideas.

Pet Project Possibility - Firefox Extensions / Redesign

The first pet project idea that I have had concerns redesigning the entire bookmark system in Firefox. The bookmarks in my opinion have not changed much from the link with name and sorted into folders. The firefox bookmark extensions currently available kind of already do the original idea I had about synchronising bookmarks with an XML doc. Therefore, I believe that I should rather than cover old ground try designing something new and innovative.

Pet Project Possibility - ToDo List App

The original idea I had which is about cross device interactivity, regards designing a ToDo List app which can store information in one place, but different devices and applications can use the info in their own manner. This has a lot to do with the flexibility or XML, which at the end of the day is just a structured text file. My original idea was to design this in .NET but now I am thinking about more java and XML since my language of expertise is Java. Furthermore, it would simplify the portability aspect for various devices as Java is excellent for running on Linux/PC/Mac etc. Oh yeh and on a mobile too.

Pet Project Possibility - ????

I am open to suggestions for ideas which are not too big. I'll list below some ideas that I have thought about : -

  • Convert a MHT file to standard HTML to view in Firefox. Implement as firefox extension
  • Make ToDo List app in XUL so it can be made as firefox extension. Like QuickNote but much better
  • Blog program / Quick Email - as a firefox extension - Nothing too complex....just like a quick emailer
  • ToDo List / Calendar using Gmail's "+ tag" - for those who don't know my email is utills@gmail.com and so if I email a message to utills+ToDoList@gmail.com then I can use this keyword as a filter to a specific label.
  • Redesign the whole preferences menu in firefox - make it more user friendly...Don't like the current design...nor the new one they've put up on Ben Goodger's site.
  • Create an internet browsing profile account....then an extension to utilise it. That way all the browsers I use can be plugged into this one file on the web, where it can retrieve my important cookies, bookmarks, and general plugins/stuff i've added to a browser. It needs to be cross browser, specifically firefox and IE compatible so that if i'm at uni its same as if I am surfing off any of my comps at home.
  • Slightly off topic.... an Algebra teaching tool...for years 8-10....basically using animation to teach key rules bout algebra. With randomly generatable Qs for them to do...that way they can basically get hours and hours of learning. Might need to do as Java Applet. or a java backend using servlets. Animation might require flash or javascript...dunno....i'll think bout this one!

If any of these exist i'd be happy to look at them and find ways to improve them.