Mindful Musings

Google does not officially support the GMail app on the Treo 755p and once you visit http://gmail.com/app with your phone on Blazer, you will receive and error message saying “not compatible with Palm OS 5 devices”. That is all true but there is a quick way to get around the problem.

There are two simple steps.

1. Download JVM for Palm OS 5 from the following link (download to your desktop computer, install through HotSync) and install it on your Treo http://www.palm.com/us/support/jvm/


2. Once that is done, visit the following URI with Blazer on your Treo, download the jar file to your phone and it will ask to install it. You should to be all set. http://gmail.com/app/v1.0.0/en/gmail-g.jar

Caveats: BUGS! The touch screen does not work very well at all on the Treo 755p with the GMail app. I am hoping that will be solved soon but the application is very navigable with the keypad on the Treo and all functions are much easier than the web version.

Enjoy!

Thanks to a Google Groups post and the helpful people on there. If you know of any other methods of tweaks, please leave a comment.

New version 1.5 now available.

I installed Mozilla FireFox 0.9 (well really an RC) recently. After a power outage I realized to my dismay that the browser would not start and would crash with an “access violation”. I traced the problem back to a preference being set (or not set as the case was). So a simple solution for people running Windows XP and Firefox is to browse to the following folder and delete the pref.js file in that folder. Restarting Firefox after the deletion should fix the issue.

Documents and Settings/computer_name/Application Data/Firefox

Pictorialis II is finally polished enough to be ready for download. The example blog can be found here:
http://mindfulmusings.net/testpict

Some of the features of Pictorialis II:
-Album building from the post interface
-Single photo addition, much like a photoblog
-Upload from FTP, point to the album, suggest an album name and description, type of thumbnail and click on publish album
EXIF support even without PHPEXIF
-GD and NetPBM support with optional NetPBM binaries
-XHTML compliant code
-Lots of custom fields for versatility and future features
-Picture preview on edit
-Picture Descriptions
-Everything is paged
-Look Ma, no tables!
-Count number of visits per picture (optional)
-Comments, trackbacks etc
-Clean simple design and CSS
-All of WP is included, so all the WP features are available!

Credits:
-Photomatt
-Unteins
-Morydd
-Jesuit
-Skippy_
-NuclearMoose
-Michel_v

Download the zip file from here:
http://weblogtoolscollection.com/b2-img/pictorialisII.zip

Please post bugs and requests on the Pictorialis Support Forums

I (we?) need some ideas for plugins and hacks for WordPress 1.2 There is already an excellent requests page on the wiki which lists some of these. If you have something specific in mind for WordPress (except for multiple blogs) please leave a comment and I will add it to the list on the wiki. I know a couple of people have been looking for “stuff to do” with WordPress and this will satisfy both parties.

I wanted to tinker with the socket capabilities of PHP and wrote some small scripts to support automatic blogging of IRC conversations. It turned out pretty well after a few changes and issues that needed to be resolved. Here are the results.

The likely suspect is the #wordpress channel on irc.freenode.net where developers, contributors as well as helpful friendly folk hang out often. All of the conversations from that channel are now blogged to http://proc.homeip.net/wordpress/ This is merely an experiment and if there are major objections to this process, this blog will be removed.

Some of the salient features of this include automatic blogging using xmlrpc. The script does not have to reside on the webhost machine. It is extremely fast and very configurable. At this time, every peice of conversation from every user is blogged as a seperate entry with the topic of the entry as the username of the person speaking. This is also turning out to be a test of extreme blogging through WordPress. The blog slows down after a certain size of the entries. All the entries are stripped of harmful tags before they are posted. Once this script has settled down a little and a few more bugs are ironed out, I will make it available for use. For reference, this uses the PHPIRC implementation for callbacks.

Ive finally settled down to a harrowing daily routine of looking for jobs and coding, all the while spending some quality time with family. Here are some of the things on my little Post-It Notes Lite on my desktop: – Fix pinging on 1.2 – passed the buck on – Checkbox for threaded comments – urgent – Recent photos pictorialis – somewhat done, needs a little fixing – Multiple upload pictorialis – Fix “who is online” on all blogs – goofy time problem on server – Write XHTMl syndication script, draw some prelim plans – Apply for jobs – throw enough shit on the wall that something has gotta stick!

Just a note of thanks to NuclearMoose (aka Craig Hartel) for creating a very nice header image for Pictorialis. That image, along with the new theme, is going to become the default distribution. I am thinking of starting another little page for Pictorialis stuff with a space for themes. Craig says that if I am desperate enough, he might come up with the first theme for it. I would also like to start a showcase for Pictorialis. I have seen some superb designs come out of it! 1000 downloads for Pictorialis and counting!!

I stumbled upon something yesterday that I thought could be really useful, at least to me. Many people have suggested and explored the idea of XHTML syndication. Some have come from big names such as Anil Dash and Tantek. However, there is very little implementation.

If every blog/CMS could have its own link for a file that spits out formatted XHTML code, life would be easier. For example, if I wanted to provide a little feed of my weblogtoolscollection, I would simply include a file from that blog that would spit out pre-formatted, pre determined code. No messing with XML, no tags to worry about or feeds to validate and then parse. The div tags etc could be defined in simple, easy to understand terms and would allow simple CSS formatting as well. A couple of different methods could be used to generate and then serve the file. Wordpress could include a synd.php (or anything else) which could be included as a php file inside another weblog which resided on the same server and had real path access. Another file called extern_synd.php (or anything else) could do a document.write to spit out the files in XHTML format inside a script tag.

Would there be any interest in this sort of thing? Here is the little peice of code which led me to this. I was writing a simple recent thumbnails for my Pictorialis and this is what I have so far. This is a simple modification of Alex King’s wp-mobile hack. I got sidetracked before I could finish it!

Anyone know how to perform a search in Google that produces unique results? In other words, if I search for weblog, is there any way that every result that is returned is from a different site by eliminating multiple results?
This seems like a very simple modification of the search scheme (to someone that does not know how the google database works).

I have made some small changes to Pictorialis which are reflected in the new download.

– Two different methods of creation of thumbnails – Choose a small area of the picture – Chose whole picture – Now has the ability of using NetPBM instead of GD, option in wp-admin/post.php – Better upload code, faster processing – Addition of extra code to weed out posts with no pictures. This blog now produces thumbnails of posts which have pictures and thus pictorialis can now be used as a full text blog as well without penalty.

There are some more updates and fixes planned, they will come as and when time permits.

I know that a lot of really technical and judicial documentation and articles have been written and published about Gmail already. However, I have felt that they lacked any insights into what would really make that service popular. Even though there are screenshots available and Matt and quite a few others have gotten to beta test that service already, GMail is still an enigma for most of us. I guess Blogger users (in random) were chosen to be given a GMail account and I seem to be one of the ones that are not chosen
In the meantime, here are some things which could be done to GMail to make it more popular to me:

1) Ability to check other email accounts built into the system. I would really like to be able to receive emails from other providers through either POP3, IMAP or HTTP mail services such as Hotmail and Yahoo. That would increase the value of GMail to me tremendously.

2) I really couldnt care less if some person at Google was reading through my email (which I am sure is not the case, just some dufus computer) but I would really like assurances that my information would not be used for anything else but ads on my email pages. If I get any notion or suspicion of any hanky panky with my personal information, I am outta there. In other words, respect the EULA, dont send me spam, protect my email address from being harvested from the Google server and do not, under any circumstances, sell my address to anyone. I dont care how many text ads you put on the interface. Period. And….ummmm..no popups please!

3) I am really looking to consolidate my email services. I do this with Outlook right now. I have all my email addresses setup to be checked on Outlook and the application is set to delete all email after 10 days on the server. This gives me a chance to get the messages downloaded on another machine before getting deleted in case something goes awefully wrong somewhere. GMail should provide this sort of an integerated experience. This thought becomes even more pertinent when considering the fact that GMail provides us with this huge amount of storage which I could use as a possible main or backup solution for my email aggregation.

4) GMail should be only about email. At least GMail should provide the configuration option to be a pure email application. I really am not interested in reading news or RSS feeds through my email application but would like to see advanced email features. I want signatures, ability to do all kinds of monkeying with my email and would like to see a vertical reading pane if possible.

5) There should be some sort of auto check facility. This could be as simple as an open browser page which refreshes every so often (configurable of course) and shows the newly received email on that page. So, instead of opening up Outlook and leaving it open all day, I could open up a GMail window and just leave that open all day. Problem solved satisfactorily.

6) SPAM filter. I would really like to see a good spam filter on GMail. I have read very little documentation on the spam filter properties of GMail but am really really happy with the Spambayes that I use with my Outlook. Something like that would be great in GMail.

7) One click book marklets. Or at least integeration of GMail into the Google toolbar. That would be awesome! If the toolbar could check your email and tell you when you have new mail, have a short preview or something, that would just baste my turkey! Also, a Mozilla compatible toolbar would be nice or I will have to rewrite all of that code to work with the present incarnation of Google in Mozilla. Bookmarklets could be used to highlight and click from a browser window to send email directly through GMail. Those are some very powerful possibilities.

Heh, last but not the least, I would like an account to test with. Matt has one and so do many other people. I really need to get lucky with that one!

More ideas coming up as and when I know some more about GMail.
PS: for those of you that still think this was an April Fool’s joke, think again.

Ping-o-matic is ready for deployment! Dougal’s Blog Service Pinger has its own domain name. Together with Matt Mullenweg, he has created a pinger for the major blog update trackers on one page. There are many major feature upgrades planned for this webpage with many cool ideas to be explored. Major enhancements in the current version:

* Your blog name and uri are stored in cookies, and automagically filled in for you when you return. * The pinger output dynamically appears in your browser as each ping completes. You no longer have to wait for all of the pings to complete before your browser loads the page. * Fixed the RubHub ping so that it actually works now. * Improved some of the error output.

Another exciting feature of this service will be its own pinging API. This site will act like a “ping aggregator” and will take care of pinging all of the other site in one fell swoop. Ping-o-matic will be the only site you ever need to ping.

There is a new version of Pictorialis out which fixes a couple of issues that were brought up with the Beta and you can download it from http://weblogtoolscollection.com/b2-img/pictorialis.zip

Fixes for this version include:

– Built in EXIF support which does not need pre-compiled EXIF in PHP – There is a selection mechanism for versions of GD which makes sure you always get the best possible thumbnails and pictures conversion. – This also has some file upload fixes which makes sure that the uploaded files do not get erased before the script gets a chance to grab the files. – The installation process has also been better defined with some fixes for the installation troubles reported from the previous version.

As in the pervious version, this new version has the following features: – XHTML compliant code – Simple once click addition of photos, auto thumbnail and resizing – Extraction of EXIF information from original photo and saved as meta – Easy installation, as easy as installing WordPress – Simple, clean design and look – Simple navigation

My example PhotoBlog with WordPress can be found at Pictorialis

On continuing my personal search for education and all things cool, michel led me to Russian Ternary computers from the turn of the century. The idea is really cool and seemed like it would work at the time. The basic principle was that instead of working with binary numbers, they worked with ternary numbers. Brousentsov’s Ternary Principle suggested that “the result of some weighing is expressed in the ternary number system by using three numerals namely: 1 (the positive unit), 0 (zero) and `1 (the negative unit).” The number base three idea follows from the original Fibonacci equations and formulae and were actively followed in Russia.
For those of you interested in this arcane yet extremely interesting computational mathematics, you can read more about it in “History of Computing in Russia and the USSR“. Apparently, Moscow University spent a great deal of money and resources looking at just this idea.
This would also make it easier to compare and come up with three pronged logic problems which require nested if and while loops (or extra temporary variables) to solve with our present implementation of binary numbers. I wish I could come up with a better example, but it completely alludes me this late at night.
Enjoy!

Six Apart recently launched their new “TypeKey” comment registration system. I have some serious problems with that system.

1) I agree with Matt and see a real problem with any centralized system that has no local component except for a closed API (more about that later). As has been discussed over and over again, any centralized spam prevention system is doomed to failure right from the start. Six Apart has many more resources than I can ever hope to have and will probably be able to handle all the traffic, but they leave the service open to all kinds of DDOS attacks from spammers which they could NOT handle. (as many other email blacklist servers and services have found out the hard way) If people say that comment spam is probably never going to reach email spam proportions, I will have to disagree. Once there is enough legislation to stop or stem email spam, blogs will just be the natural step down for spammers. But that is an arguement for another day.

2) I dislike that company’s elitist propaganda. Here are some excerpts from this TypeKey page:

“And TypeKey will work with TypePad weblogs as well as other tools and services soon after that.”

They expressly state that there will be a product “launch” and then …

“At that point, there will also be information about what is required to make use of TypeKey services in commercial applications.”

Six Apart is a growing company with many talented individuals who have defined blogging as we know it today. BUT, they are not Gods and never will be! Microsoft-esque promotional documentation language is just annoying and feels very elitist. Just for that I refuse to use their service.

3) How is typekey different from forcing users to register on a blog and only then allowing them to comment? What if I sign up for typekey and then post all kinds of junk?? who says whether my stuff is spam or not?? Some big shot blogger might think that my opinion on their “learned” topic is spam and decide to report me to TypeKey. Would that mean I could not comment anymore? Is one person’s comment another person’s spam?

4) This is proprietary software folks. Nothing to see here. There will …

” ..... be information about what is required to make use of TypeKey services in commercial applications”.

5) What about trackback spam? How will TypeKey prevent trackback spam? Get rid of trackbacks completely? Is there another component? No solution there at all.

6) As Michel points out, what about identity theft? What if someone signs up as Mark Pilgrim with a different email address and posts all kinds of junk? Who is to say that Mark’s email address is blah@blah.com and not bah@blah.com? Centralized solutions of establishing identity break down without getting really personal information involved. I better make sure I am the first to sign up with all my names and nicknames!!

“TypeKey helps ensure that people who comment on a site have a verified identity, keeping conversations on track and helping to prevent abusive or offensive content (comment spam) from being posted.”

– Nah! Doesnt work right and I aint givin you my credit number neither!!

With cautious optimism and glee, I would like to say that I might have more time to dedicate to a job search and coding now.
As for the items on my plate, the more important ones are:

Promoting my Peace and Unity Meme (see above, please)

Applying for jobs for at least 2 hours a day (I have to be regimented, or I lose track)

Writing a very simple MySql based wiki’ing system which is designed to be intgrated into other systems (primarily WordPress)

Fixing my threaded comments hack to be compliant and non-javascript (and other little issues with it)

Writing some code to support a LiveJournal mod of WP which would allow more control over the posts (might take a while)

And of course, cooking and spending time with my Jennifer

My parents are going to here on the 12th of April and that could not happen a day too soon! I have not seen them for almost a year and a half!! Jennifer Luckey will get sooo spoilt (again) !

What does the email author and the email receiver have in common that the SPAMmer does not? They are related by some sort of a social network. Leaving the obvious examples aside, if someone from the WordPress or the Drupal forums wanted to email me, they would already be a part of my social network.
This process could be used to check all email. You could have a web interface (much like orkut) where you setup your email accounts to be checked through one interface. Once they are setup, instead of whitelisting or blacklisting email addresses, you would add or delete someone from your own social network. In order for an email to be accepted for delivery, the social connection would have to be developed first. This could be as simple as adding someone to your own list of social contacts once you receive an email and have not categorized it as spam. You could receive emails from people in networks which are close to yours such as your co-worker or friends. The power of this method is hidden in the fact that no-one really wants a spammer to send them email. So even if a spammer wants to get invited into a network and start spamming, they would be kicked out very soon by the users.

Look at the way you read your email. Wouldn’t you rather read the messages from your closest group of people than those from a diamond miner in Uganda? So this system would arrange emails according to your closest circle outwards. The last circle consisting of completely unknown people who are probably sending you spam anyways. As you delete these messages, you tag them to be removed from your network and add them for a master blacklist of emails to be NOT allowed to join these networks.

The recently came across a web based email system that allows for their emails to be recalled after being sent, to be viewed only once and to be accounted for after they have been sent. That concept has always been very interesting to me and there have been many instances where I would have loved to be able to perform thos tasks with my own email. That led me to orkut which allows users to send emails to groups. I find that most of these emails from groups are what I consider SPAM. If I had a way of notifying the system that those messages are spam and the system had a way of warning the user sending those emails (and finally removing them from the group) the spam issue COULD be brought under control (in orkut).

That led me to thinking about social emails in general and how social networks work in particular. Who would want a spammer in their social network? If all incoming and outgoing emails of a society are controlled by a central authority, that central authority would have a much better understanding of all the emails sent and received. If the policing of SPAM is done using constructs of social networking, would human engineering work to our advantage? Could there be easy ways to break the system?

I would really like to work on setting up such a system for testing. However, I sometimes think that my own ideas a really off the wall and tend to shy away from making them a reality. Would anyone be interested in discussing this with me? Do you think this would work? Where is Orkut when I need him!

Here is a link to Matt’s post about “Code is Food“.
I absolutely adore the “one more won’t break the web” analogy coming from some of the people who have left comments on that post! That embodies the difference between people willing to get off their asses to make a difference and people trying to justify the problem and ignoring the issue at hand. Sweeping the problem under the rug because all the cool kids are doing it” is not an acceptable answer for a programmer, technologist and/or a web solution developer.
This post represents Matt’s opinion. He believes in clean, compliant code and works dilligently towards promoting the cleanliness of code structure and most of his work is representative of the same. My refute to the people against Matt’s views is that, LockerGnome has a large viewership and should be on the forefront of web technology, embracing rather than deprecating the nuances of proper web coding procedures.
There were some other really interesting comments on that post, some coming from people that I had a fair amount of respect for in the past. Some wanted to rewrite WordPress in Java (I say, go for it!). Others expressed their derision of WordPress and PHP. However, the most interesting comments are the ones from people actually standing up for LockerGnome and the lack of web standards in the new design.

All of that having been said, I will blushingly add that my own code on this blog is not standards compliant. I have no interest in staying up late nights to make it standards compliant. If it happens, it will happen in its own sweet time. However, I also do not run a website that is considered to be one of the fore runners of computing technology, nor am I that famous or important.
I know … excuses, excuses … maybe once this thesis is done.

I am done with most of the work for WuhWuh 0.1 Release Candidate 1
It can be downloaded here and the forums are located here.

Simple installation, lots of features and a clear upgrade/migrate path for existing WordPress users are some of the salient features. (this is basically a hacker version of WordPress 1.2)
See you on the forums!

PS: If you like it, I would appreciate it if you could publicize it somehow on your blog. I did not put any links on the installation and will leave that upto you. Thanks!

I have put together a “hacked version” of WordPress 1.1 alpha and would like to get some feedback on it before I package it for release. Here is the link to “WuhWuh” version 0.1 (pronounced woo’aah woo’aah)
Some of the prominent features include simple 3 step installation and a built in theme manager.

Username for login: admin
Password: password

I have a long functional document (explaining all the hacks, functions and ways to use them) along with a wish list and future enhancements. If this does catch on, expect a seperate set of forums (to prevent confusion the original WP forums) and a release cycle of once per major release of WordPress.

And ummm…people…dont change the password! That is just stupid to do. It is my database, my server, my control. It just makes life a little more difficult for other people trying to log on. So…ummm….grow up!

I have the proof of concept for my thesis relatively completed. I realize that it need not be perfect, but I wanted to get it functional. It features a shopping cart for a Coffee Store and lets the user pay with a “virtual” card which emulates the algorithm for my thesis. I had to rewrite a full shopping cart script and decided to include paypal support for anyone interested in using the code. At this point in time, the user has to click on a link called “click here to see smart card” to get the smart card simulation running. The rest explains itself. Here is the link to the page if anyone is interested.

I still have to finish writing the second draft for my introduction, get some passport photographs taken at Walgreens, start doing my laundry (thankfully, I have my own machines) and finish the ER diagrams and the writeup for the Concept shopping cart. (I might make some ER diagrams for WP while I am at it :)) Thats the plan for today. Jennifer is working at Rite Aid today and will come over late in the evening when I talk to my parents, which should be followed by dinner and another couple of hours worth of work.

Surprisingly, this weekend has been amazingly productive. Our thesis deadline has been extended (does not really matter, I am going to finish according to my deadlines anyway). I am working with WordPress whenever I have some free time. Look for some really interesting changes in the new version, along with increased stability, increased character encoding compatibility, better install and (from more sources) import algorithms.

Luckey feels neglected with me being this busy, but she will be in heaven in a month (my parents are going to be here then)

Here are some examples:
Here is a WordPress 1.1 beta blog showing 300 posts.
Here is the same blog, cached with my method

Take a quick look at the bottom of the page for the generation times. The number in small letters in the center is for the main entries to be generated and displayed and the larger number to the left is the time for the whole blog to be generated.

A little more work is required to make this a functional hack. The categories and comments sections (along with archives and others) do not use the cache at this point to display the pages. Only the index page uses the cached entries and produces all the right links. Consequently, some more work is required to update the cache when some information is edited or added. This could include adding or editing a post or addition of comments and/or trackbacks. However, with a larger numbers of posts and links, this cache speeds things up quite a bit from what I can tell. I had to mess around with quite a few of the functions to get them to work correctly with storing a variable instead of echoing them. If someone is interested in the code, leave a comment or email me.

What do you think? Is this something worth investigating further?

In my opinion, the part of WP which takes up the most MySql queries (and the most time responding to a request) is in setting up the posts for display. If there was another table in which the complete post markup was stored (including the post title, author, time, number of comments etc, one single entity created with all the HTML markup, and the filters are executed on the post) after being created (with the postID as the key for the table), the posts could be loaded multiple times relatively quickly. When a comment gets added to a post or a post is edited, the whole row is deleted (in the new table) and recreated with the same postID and new markup that is generated. The blog-header could contain code which would allow a complete rebuild through a switch. There might be the need for a second blog-header to manage the two different types of display requests.

The categories and archives would work in very much the same way. The blog-header that controls the static posts would do a select postID from all posts in the main posts table for a category and then select again the same IDs from the statis posts table and display those in the order chosen. Permalinks could work the asme way they do right now with comments being selected live from the database etc. Rebuilding (adding or editing one post really) the static post table would take place at the end of a post or edit process and should be quick and painless.

Any comments on this?

I have been trying to work very diligently on my thesis and am starting to make some progress. However, there is considerable amount of work ahead of me in the next couple of weeks. My friends are being ignored for the time being. I do have to thank Jennifer, Al and Luckey for being as supportive as they are (and of course my parents for encouragement)!
As a direct result of my “diligence”, as I so honestly put it, I have been ignoring my code. Here is a small list of items that has come up during that time.

Stuff to Do: – Threaded comments rework and extend functionality – Cached technorati – Sticky post debug – Import/export blacklist – Think about a way to do static pages for WP

On that last point, here are some questions I have (for the MySql gurus, namely JeremyZ, PhotoMatt, AlexKing, TechGnome and others) . I am looking for a way to tell the last INSERT or UDPATE that was performed on a MySql table without having to keep track of every INSERT and/or UPDATE. I would also like to be able to count the number of INSERTs and UPDATEs performed on a given table. Are there any housekeeping commands that MySql provides besides SHOW STATUS ? Can SHOW STATUS be used on a table?

#EDIT# Found the answer!

SHOW TABLE STATUS [FROM db_name] [LIKE wild]
SHOW TABLE STATUS (new in Version 3.23) works likes SHOW TABLE, but provides a lot of information about each table.