Isles Tech

Since May 1, 2008, exactly at midnight Philippine time, I inserted a StatCounter code on Isles Tech to give me a way to tell the current visitor how many unique visitors had been here and how many pages had been served. Two visits from the same IP address are not considered unique if they happened within 30 minutes of each other.

I'm still keeping the Google Analytics tracking script which I've installed here a few days ago. (I don't know how to access via a script the information from a Google Analytics report.)

I'm using a blocking cookie to prevent my own visits from being counted.

Like this post? Please help me pay for my Internet access.
Links to this post
Labels: google analytics, statcounter, statistics counter, tracking script

This blog started as a collection of notes I had written for myself since I started learning how to program on the computer. When I got hooked on the Internet I thought it would be a good idea to transform the notes into articles in a blog, although the actual conception of the idea was separated from each other for a good deal of time. I jumped from Blogger to Wordpress.com to Blogger again, and now I'm fine with this.

The good thing I noticed with this blog is that I don't need to remember where exactly did I place some note; the search function is there. So there's really no effort at coming up with a navigation, and labeling is just rudimentary. (However it should be noted that it took me some time to accustom myself to this setup.)

While I was riding the jeepney towards my hometown this evening, I thought it would be a good idea to use Blogger.com for a website which is more traditional and static, that is, a website which is not a blog. One of the things I'd like to implement would be to remove the date information from the Blogger URL; apparently that is not possible. As for navigation, I thought it would be best to just use the search function, apparently (again!) it seems there are (still?) more navigation- than search-based visitors.

So are navigational aids in a Blogger.com website (whether blog or not) still necessary? It would seem so.

Like this post? Please help me pay for my Internet access.
Links to this post
Labels: navigation, searching, usability, user help

There seems to be no way of removing the date information from the URL of a Blogger post. If that was possible, it would be very easy to create a website on Blogger which would not look like a blog, essentially making Blogger.com into a webhosting service.

Like this post? Please help me pay for my Internet access.
Links to this post
Labels: blogger as webhost, blogger URL, blogger website, blogger.com

I was reading through Milan Parmar's post on how to get paid by Google for user research studies and (surprisingly for me) clicked on the link for an application. (I'm not really into applying for things left and right; just consider that this blog already had more than 200 posts before I decided to step up things by applying for Entrecard, BlogRush (failed), BlogCatalog, Adsense, etc.)

The survey I took (just for the application) is worth it, even if my application would be denied. I was even surprised with my answers. For instance, I realized that I'm not really into photo sites like Flickr.com and PicasaWeb (I did not even know its URL until now :-)), and although I frequent YouTube, I don't know how to create even a simple movie. Surveys like this get you into thinking about your daily online habits :-)

EDIT: Here's a longer review.

Like this post? Please help me pay for my Internet access.
Links to this post
Labels: Google, user surveys, webreview-milan

After I decided to install a chatbox here at Isles Tech, I noticed that the first two commenters (my girlfriend and an old friend back in college) used their email addresses. The Cbox.ws chatbox explicitly tell the potential commenter to use either his/her email or the URL for his/her website. Since a good percentage of readers do not have their own blog or website, they will use their email address, which are then converted to mailto: links. The problem with this setup, though, is that spammers all over the world run scripts which look for mailto: links. And if such link to your email address is seen, let your mailbox welcome thousands of unsolicited commercial endorsements. (A note: I don't think this could actually happen with the Cbox.ws service, although there seems to be no way of disabling automatic conversion of email address to mailto: links from the Cbox.ws management panel.)

This is where the use of Friendster profile URL will be useful. Who does not have a Friendster profile (at least in the Philippines) these days? So instead of providing one's email address (and thus be open to spam), one could just provide a link to one's Friendster profile. (If one does not have a Friendster account, then the URL to someone's Multiply or even Facebook account could be used.)

This is also useful for bloggers who want to create a link to some people... who don't have blogs or websites. Who is Mr. Gone? Who is Mr. Cagurol? They're my co-teachers in school but they don't have their own websites. How am I supposed to provide a link to more information about them? A page about them on the school's website would be very good, but as of this writing there is none. A note though: use the system-generated profile address (those with string of numbers after http://profiles.friendster.com/) instead of the personalized URL of the Friendster member, since the personalized URL may be changed by the user.

Like this post? Please help me pay for my Internet access.
Links to this post
Labels: chatbox, facebook.com, Friendster, mailto:, multiply.com, spam

Openomy.com is an online file storage system. You can use Openomy as a backup server for your files, up to 1 GB. The organization is via tags (which I hate) and not by folder. The file manager is very basic, but I like it for its simplicity. However the site seems sluggish: uploading a single 8KB file took almost a minute! This was even after they explained something about DDOS' attack.

The cool thing about Openomy.com for programmers is that they have an open API, so you can write programs accessing the service.

By default your files are kept private, but you can provide a public URL for your files; my ubiquitous avatar at the right, for example, can be accessed at http://files.openomy.com/public/islesv/islesv-128.jpg.

If not for the sluggishness of the site, I would have recommended it.

Like this post? Please help me pay for my Internet access.
Links to this post
Labels: file hosts, openomy.com

Since an ampersand (&) followed by three of four characters and then a semicolon is interpreted as special by web browsers, the way to write   (the character for non-breaking space), the proper way to write   would be   (can you imagine how to write the previous one? :))

Like this post? Please help me pay for my Internet access.
Links to this post
Labels: ampersand, HTML, special characters

http://www.beam.org.ph/, 4/1/08:


In the tradition of Microsoft's 1 files(s) found :)

Like this post? Please help me pay for my Internet access.
Links to this post
Labels: beam.org.ph, web opps

From NG Khai's post on the NK Educ Summit 2008 last April 23 (when I received the Google Alerts for "education" and this showed up):


If you to the page now, the first and the last   are already gone, but there is still the   after the third sentence. In addition, the pictures are gone:



Like this post? Please help me pay for my Internet access.
Links to this post
Labels: education summit, Microsoft, ngkhai.net, web opps

I've seen this before but did not bother getting a screenshot. Now I have one. Take note of the login form:


Like this post? Please help me pay for my Internet access.
Links to this post
Labels: GMail, web oops

I've decided to merge my Web Oops! blog with this one. Web Oops! is supposed to chronicle web errors which I stumble upon now and then, but it barely got off the ground since I started it last March 31, 2008. That's the main reason why I'm just merging that blog with this one.

For those of you whose focus are on technical issues, that should not be a problem. Just ignore those posts :)

Like this post? Please help me pay for my Internet access.
Links to this post
Labels: web oops

Say, where should I click? :)



Like this post? Please help me pay for my Internet access.
Links to this post
Labels: blogger.com, web opps

Most webmasters and bloggers put up a shoutbox somewhere in their pages. Blog-Tutorials.com recommends the use of a shoutbox to increase productivity in a blog. A shoutbox gives visitors a chance to inform the webmaster or the blogger that they were there; while commenting on a post takes some effort, putting a smiley or a simple "hi" or "just blog hopping" is certainly near to effortless. (However, take note that this can be abused.)

The shoutbox also provides a way for users to interact with each other in general terms. Since there is no "registration" it is very easy to post a message.

I've decided to put up on this blog my own copy of the Cbox widget. Hopefully visitors will be commenting. (Just a side note, most of my personal friends are not into technical things, except possible for Jo Alton and Edick, so I don't expect much.)

I chose Cbox over TabBoard, Free Tagboard, TagBox, and ShoutMix because of familiarity - I already had an account with them years ago. The free version of Cbox limits the number of messages that are archived to 100, and one email can be associated with only one Cbox, but there is nothing in the terms of service which states that you can't create multiple accounts.

Like this post? Please help me pay for my Internet access.
Links to this post
Labels: blog interactivity, cbox, chatbox, shoutbox

Two days after I've installed Google Analytics (see How to track your Blogger Statistics using Google Analytics), it almost confirmed my guess that this blog is getting at least 30 unique visitors everyday:



Like this post? Please help me pay for my Internet access.
Links to this post
Labels: blog statistics, google analytics

List of things to keep you motivated to blog (thanks, Michael Aulia):

1. Blog out of passion, not for the money
2. Blog the thing that you are familiar and passionate about
3. Install or use some analytic tools to your blog
4. "Even if there is [only] one person reading my post, it's still worth it"
5. Don't let the criticism bring you down
6. Visit other blogs and if you like the post, comment on them
7. Connect with other bloggers/people

You must read the full article to really appreciate it.

Like this post? Please help me pay for my Internet access.
Links to this post
Labels: blog, blogging motivation, motivation

Out of the top 50 of the Top Philippine Websites, more than 10% are of adult nature. iyotTube.com is at no. 11, while the less pornographic (but still lewd) FHM Philippines is at no. 22. There are two sites the name of which do not indicate anything pornographic, yet they are anything but: txtflix.com (no. 28) and jologs.com (no. 29, with its legalistic but probably indefensible disclaimer policy). There is also the "sexy story" collection at pantasya.com and the more-or-less adult forum at manilatonight.com

On the other hand, the name pinoyunderground.com (no. 19) suggests something, but because the administrator requires people to register before they are given access (a bad idea), I was not able to see what's inside. Also, katambay.com (no. 36) told me that there is a problem with the bandwidth at the moment. :( nyakers.org (below 50th position, no. 57) also have bandwidth problems.

Flesh Asia Daily technically does not deal only with Filipinas (not safe for work), but it is included at the Top Philippine Sites at no. 74.

There's seems to be a profusion of porn wannabes of YouTube. Aside from iyotTube, you also have PinayTube (no. 59 and 60 - there is a problem with the database) and Pinoy Scandal Blog (no. 70) - perhaps I should just create my own too using StartYourTube.com (but...).

(My brother knows of at least one other adult video site catering to "Pinay scandals" which is not in the first 100 sites of Philippine Top Sites - Pinay Scandalism. And wasn't it just last week when the beautiful Sen. Loren Legarda urged cybercrime units of the government (are there any? :-)) to shut down The Pinoy Tube? BTW, here's Mark "BoyBastos.com" Verso's take on that issue.)

With these kind of sites proliferating in Philippine cyberspace, I don't know what's coming to Filipino internet.

EDIT: PutangInaMo.com (no. 260) also has a video sharing section.

UPDATE: Libug.com and FreePinayScandal.com also contain this kind of thing. (NBI, thank me. :))

UPDATE 5/5/08: The Pinoy Tube had already been shutdown:



Like this post? Please help me pay for my Internet access.
Links to this post
Labels: philippine websites, pinay scandal

Say again, you're selling what?



Like this post? Please help me pay for my Internet access.
Links to this post
Labels: web opps

As I thought over my problem of having PhilTowns Search always including sites belonging to the PBwiki network (that is, any site having pbwiki.com in their name), I've decided that the best workaround for me at the moment would be to have a site-wide exclusion for PBwiki (sorry PBwiki fans). I think that there might be good resources on Philippine local history and genealogical studies on PBwiki, in fact, now I could not include the archives of Bisaya Magasin, not to speak of my forthcoming TownpediaPH, but I felt that I did not have a choice since anyway I would have to create an ExcludePartialSites page, and that page would certainly contain a link to http://pbwiki.com/privacy.php which would then exclude all PBwiki sites.

Like this post? Please help me pay for my Internet access.
Links to this post
Labels: pbwiki.com, PhilTowns Search

I finally discovered why all PBwiki sites are being included in my PhilTowns Search Engine. Using the Get URLs tool from paulding.net, I discovered that http://pbwiki.com/privacy.php is one of the links in my Directory page (from where the links are dynamically extracted). Since I chose 'partial sites' inclusion, all sites on the level of pbwiki.com and below (which is to say all subdomains of pbwiki.com) are fair game for the search engine.

Now that I have seen that I think it will be more logical for me to have a PagesDirectory, PartialSitesDirectory (subdomain.domain.tld) and SitesDirectory (domain.tld), and follow the appropriate choice for dynamic extraction of links. (And this also gives me another idea for a user-based search engine. :))

But for now it appears to me that with the current method there is no easy way for me to limit the search to only PBwiki sites which I have explicitly included; there will always be the partial sites inclusion of http://pbwiki.com/privacy.php :(

Like this post? Please help me pay for my Internet access.
Links to this post
Labels: pbwiki, PhilTowns Search

Instead of specifying manually the sites or pages to be included in a Google Custom Search Engine, you can opt to have the CSE extract dynamically the links from a page:


From here, you have three choices (thanks googlesytem.blogspot.com):

1. Include only the pages the page link to
2. Include the partial site of the linked page. A partial site is indicated by a subdomain. E.g., if the page contains a link to http://tech.islesv.net/2008/04/startyourtubecom.html the partial site would be http://tech.islesv.net/ and if you choose this option, the pages under tech.islesv.net would be included in the search (but in my experience, those of http://blog.islesv.net/ as well).
3. Include the entire site, as indicated by the domain.

I've done some testing and it seems to me that this method is still a bit unstable; the same search term sometimes give me two, one or zero result for the same phrase. Also, there doesn't seem to be no difference between the partial site search and the full site search.

Like this post? Please help me pay for my Internet access.
Links to this post
Labels: dynamic links, google cse, link extraction

This is a normal PBwiki page:


And this the same page with ?raw=bare appended:



Like this post? Please help me pay for my Internet access.
Links to this post
Labels: pbwiki

PhilTowns Search is a search engine which uses the power of Google CSE to provide users the ability to find specific information about Philippine towns in a time-efficient manner. When you use Google nowadays to search for terms related to Philippine towns, you are presented with a lot of unrelated information. For example, for Pinamungajan, Cebu you will find entries related to mp3 music downloads or even a comment on the current rice crisis - on the first page of results. PhilTowns Search provides an interface to Google which prevents these irrelevant results.

How is this done? We utilize a set of reviewed sites and Google Custom Search Engine which extracts links dynamically from webpages. Essentially, only those pages, parts of sites, or entire sites which we have already reviewed to be worthy of inclusion in the directory are actually searched by Google when you do a search. There is a positive side effect of this process: some sites which have too few links to appear in the first pages of results when using just Google will be seen if they are in the directory.

Sites which are not included in the database will not appear in the results. If you are a webmaster, please take a look at our directory and see if your site is listed. If not, you can contact us. Please don't forget to include the URL to your site. You can also use the form for other concerns.

Please visit PhilTowns Search now and explore its capabilities.

Like this post? Please help me pay for my Internet access.
Links to this post
Labels: local history, philippine town search, philippine towns

Google Custom Search Engine allows you to specify a page (or pages) where you can put links which are then extracted dynamically by Google itself and added to your search engine. Very useful for my PhilTowns Search project which is a search engine for sites, partial sites, and pages related to Philippine towns, local history, and genealogical information.

I might have missed this functionality the first time. (It is marked "New", but so is the Custom Search Help Center.) I did a search at the Custom Search Help Center but the closest I got was a query last April 11 - which appears to be answered by this functionality.

Number one preferred site for my PhilTowns Search is of course my very own Townpedia PH, which is itself an encyclopedia of Philippine towns, local history, and genealogical information.

Like this post? Please help me pay for my Internet access.
Links to this post
Labels: dynamic links, google cse

I just came across an interesting site known as StartYourTube.com, which has a very good catchphrase: "YouTube sold for $1.6 billion. How much will you sell yours for?". Looking through the website, and being tempted to create my own tube right away (despite not being even an active bystander at YouTube - I rarely watch videos myself), I was surprised that they don't seem to have a viable business model. They're even giving out $500 every month for the top "tubes".

Okay, so you can create your own YouTube-style site, you can customize its colors and logo, you can even mask it with your own domain. What's in it for the creators of StartYourTube? They don't even get a portion of your ad revenue, should you choose to put ads on your site.

With the opening up of the YouTube API I wouldn't be surprised if StartYourTube.com wouldn't see the end of this year.

Like this post? Please help me pay for my Internet access.
Links to this post
Labels: business model, startups, youtube

110MB (how to), Blogger (how to), WebNode (how to), Google Pages (how to), Weebly (how to), Tumblr.com (how to), FreeWebs (how to), more.

Like this post? Please help me pay for my Internet access.
Links to this post
Labels: free web host

1. You can't put links in the footer.
2. The forum widget is not a fully functional forum.
3. It seems WebNode only support second level domain names, i.e. only domain names of the form something.tld (islesv.net); subdomain.something.tld (townhist.islesv.net) is not supported.
   

Like this post? Please help me pay for my Internet access.
Links to this post
Labels: webnode, webnode.com

dot - used to override IDs. sharp - used to override classes.

Like this post? Please help me pay for my Internet access.
Links to this post
Labels: css

WebNode is a free service which allows one to design websites using drag-and-drop. The interface is very easy and intuitive. There are hiccups in the system, of course, but generally they are ignorable compared to the superb service the system gives you.

The templates provided in the webnode website has the string "Create free website with Webnode" somewhere near the footer.

You can hide this string by taking note what is the id of the div it is contained in, and then putting the following HTML in the footer itself:

<style>#footerDwn { display:none; !important }</style>

If the div's id is not footerDwn, replace accordingly.

I haven't seen anything against it in WebNode's Terms and Conditions, but I could be wrong.

Like this post? Please help me pay for my Internet access.
Links to this post
Labels: hiding strings, webnode

Less than a day after taking a jab at bloggers who focus on number of feed subscribers, I give in to the pressure and added my own:



Like this post? Please help me pay for my Internet access.
Links to this post
Labels: feed subscriptions

My project to create a free Philippine law database was originally called Lex Libera. When I realized that there might be a potential conflict with the established Lex Libris of CD Asia Technologies, Inc. I renamed it as The Free Law Project (if you click that link, you will be taken to EduPhil.org's main site, because I'm too lazy as of the moment to create the necessary routing links). The project is currently in hiatus, as I find other things worthier to pursue, but definitely I'm not stopping that project. Meanwhile, I spend more time with Lexpedia, a free Philippine law encyclopedia for nonlawyers.

The original Lex Libera project was registered at wiki-site.com. That was before I chose to find a reliable free web host which offers PHP and MySQL, thus allowing me to run MediaWiki, which was my platform of choice. As wiki-site.com does not have mechanism for automatic deletion of your wiki site, I had to leave a message to the administrator:


Surprisingly, until now the site is still active and receiving a few hits now and then:



Like this post? Please help me pay for my Internet access.
Links to this post
Labels: freelaw, lex libera, philippine law database, wiki-site.com

By default PBwiki will only allow scripts inserted by the wiki's author(s) to run if the user allows (here's why). The procedure to allow this requires clicking on an "Enable scripts" link at the bottom of the page. That link is dynamically created by PBwiki and you can't just put a link in your page saying, "If you can't use our functionality, you have JavaScript disable for this page; please click here." And although it might be possible to enable JavaScript permanently on your PBwiki by emailing support, I don't like to do that because I have a couple of PBwikis that I might need to have the functionality enabled. So, I created a workaround.

The workaround I am referring to had been applied in my Lexpedia wiki. That wiki utilizes Google CSE (and hid the default PBwiki search bar in the process). The search bar is placed on the QuickStart at the side bar; I did notice though that I was not prompted to enable scripts here, yet apparently the Google CSE watermark (which is a JavaScript effect) is applied:


(Does this mean that scripts in the sidebar are always enabled? Might be worth an investigation.)

The results page contain the following text:

If you do not see any search output after the line below, you need to do the following steps:
1. Click here to reload the page.
2. Look for an "Enable scripts" link below and click on it.
3. Press Ctrl and F5 at the same time (bypass your browser's cache)
4. Reenter your search term(s) in the box at the right.

Step 1: As you can see, we provided a link to the page http://lexpedia.pbwiki.com/Results - the reason for this is that the actual URL opened after a search is done contains other parameters, e.g. http://lexpedia.pbwiki.com/Results?cof=FORID%3A11&q=contract&sa=++Search+Lexpedia++&cx=004397709276413070647%3Aebv9m_yqo9e . Consequently, the ?scriptkey=xxxx&set=1 code is appended to this long URL, which breaks up things. What we need to get is http://lexpedia.pbwiki.com/Results?scriptkey=xxxx&set=1 thus the first step.

Step 2: As noted above, the actual URL link is created dynamically; the xxxx in
?scriptkey=xxxx&set=1 above will be replaced with a code. So we have to instruct the user to find the link for him/herself.

Step 3: In our test runs we never had succeeded in getting a result on the first try after enabling the scripts, and bypassing the browser's cache seem to be a solution.

Step 4: We have to request the user to reenter the term(s) he/she was searching for.


Like this post? Please help me pay for my Internet access.
Links to this post
Labels: javascript, pbwiki

It seems you can just ignore WebNode's warning regarding the record "files".

I am using EditDNS.net, and I have the following records:

www.islesv.net. CNAME 1 86400 islesv.webnode.com.
files.islesv.net. CNAME 1 86400 islesv.webnode.com.

(Since EditDNS.net cannot handle CNAME entries for root domains, I'm using URL forwarding for islesv.net to www.islesv.net)

But when you go to your domain manager at WebNode, it will tell you that the record for name files is not set.

However, every thing seems alright when you open the site. For now, I would continue my work. :)

EDIT: The problem will be apparent once you start inserting images into your pages. The images will all be sourced from files.username.webnode.com, but you don't have that, so the picture won't show.

Like this post? Please help me pay for my Internet access.
Links to this post
Labels: record "files", webnode

It was not until this afternoon when I tried using domaintools.com's whois service that I noticed that http://www.islesv.net/ is not redirecting to http://islesv.110mb.com/i/ unlike http://islesv.net

This is what one will get if you type http://www.islesv.net/ :

In a previous post I noted that Blogger.com can very well be used as your webhost, and all you need to pay for would be your domain name (since Blogger is free). The only caveat is that Blogger is a blogging platform first and foremost, and therefore your entries will always be in the "time stamped" URL. Nevertheless you should not look at that as a complete problem: the "time stamped" URL then implicitly conveys the time when you started creating your article (by default not the date you actually published it; see here).

Some (all?) Blogger templates allow you to drag text page elements, which you can use to store static content on all pages of your blog; this is where you can put the request for the user to subscribe to your feed, but see why this might be bad.

Conceptually, http://www.islesv.net/ should be the same as http://islesv.net/ - this is the default understanding of many people. Unfortunately, EditDNS.net (my free domain manager) can't handle CNAME entries for root domains, and pointers to Blogger are handled as CNAMEs. Therefore I can't have islesv.net pointing to some blog at Blogger; instead, what I need to do would be similar to the one I've done for bisdak.org: URL forwarding for bisdak.org to www.bisdak.org, with the www.bisdak.org pointing to a BlogSpot blog using CNAME.

The same notes above are applicable to WebNode, which I will review later.

Like this post? Please help me pay for my Internet access.
Links to this post
Labels: blogger posts, CNAME, domain name, editdns.org, URL forwarding, webnode

I've always loved PBwiki since the time I stumbled upon it while searching for products similar to TiddlyWiki. I have a lot of wikis in their site, most of whom are private.

I've taken a look at their whitepapers and although I already know that PBwiki is the best solution for my personal wiki needs, I was convinced that it could also be the best solution for our school's internal collaboration requirement. Briefly stated, instead of paying for technical support, the school could outsource the technical requirements to PBwiki.com

I tried to propagate adoption of an internal wiki last year, but it did not take off the ground. We were using MediaWiki and there were a lot of issues: security (accounts had to be created manually since students also have access to the intranet when they are in the computer lab, although I doubt if any of them knew the address), technical misunderstanding (I was not given access to the router, and people had to access the system via its numeric IP address, which slowed down conversion), and the learning curve required to know Mediawiki code.

This time around, I hope to convince management the usability of running a pilot test of using PBwiki.com

Like this post? Please help me pay for my Internet access.
Links to this post
Labels: collaboration tool, pbwiki

I'm starting a collection of links to my own posts and other people's posts which deal with the cheap ways some bloggers use to lure readers into their blogs.

I'm not keen on starting yet another blog for this, but we'll see.

Like this post? Please help me pay for my Internet access.
Links to this post
Labels: blog readers advocate

1. Write something about Pinay scandals, cellphone scandals, etc.
2. Use a graphic for your blog feed subscribers to trick innocent users in signing up - after all, 28,000 readers can't be wrong.

Do these two things and you're a cheap blogger.

Like this post? Please help me pay for my Internet access.
Links to this post
Labels: blog baits

I can't understand the emphasis some bloggers place on increasing the number of subscribers to their blog's feeds. If you are not short of forced to register for someone's blog feed, the blogger use baiting techniques to lure more subscribers or uses a graphic which says 28,000 or something readers.

I still put a subscribe link somewhere on this blog to make it convenient for those who would like to read the latest posts using their own readers. However, I do not make it a point to force users to subscribe, since I sincerely doubt that all my posts would be useful for any single person. Heck, sometimes I even find posts which I never thought existed in my blog, and which I could not readily remember for what purpose I posted, for example, my diatribes against Globe Broadband (here) and PCQuickBuys (here).

Realistically speaking, increasing the number of feed subscribers is more for the ego of the blogger than for the benefit of the subscriber. If someone really want to read some parts of my blog (and I thank you for that) it doesn't follow that he would want to be bothered by my next posts. After all, what do all those numbers really mean? I am increasing by one some people's number of subscribers, but it had been months since I last visited my online RSS aggregator BlogLines.com - just because you have X number of subscribers doesn't mean that all your new posts are read by X number of readers.

So next time that you stumble upon a blog which insists that you subscribe to its feed before you access any of its content, run. That blogger is not worth your time.

(P.S. I have plans of creating "the most-subscribed to Cebuano blog".)

Like this post? Please help me pay for my Internet access.
Links to this post
Labels: blog feeds

For a minor blogger like me, paying for annual registration fees for domain names can be daunting in terms of the financial strain it imposes.

As of the moment, I have four domain names registered under my name: islesv.net, for my personal site; dilfed.org, supposed to be for The DILFED Forum, as yet undeveloped and just redirects to the group's Yahoo page; bisdak.org, another domain name which is not yet fully developed by me - I just registered it to prevent other people from registering it; and philtowns.net, perhaps the greatest failure I ever had online.

Each of these domain names I have to (re)register for 600 pesos every year. That's an equivalent of five to six days at the office.

I had resolved to myself stop from buying domain names unless I have very good reasons to do so (for example, in the case of bisdak.org). This is why philtowns.net was a bad one, because I did not do a thorough market study before I worked on it. (Say charge it to experience - at least that was the first domain name I registered myself, and I learned a lot in the process.) For those interests that I have at the moment, I will just use a subdomain of islesv.net

Like this post? Please help me pay for my Internet access.
Links to this post
Labels: domain names

I just figured out that you can use blogger.com as your webhost. The idea is to buy a domain name (islesv.net in my case) and use different blogs on Blogspot for your different interests (example: this one is at http://tech.islesv.net/ which is an alias for http://islestech.blogspot.com/ while http://blog.islesv.net/ , my teaching blog, is an alias for http://islesv.blogspot.com/ )

Of course I've seen a lot of other people already using Blogger.com as their (free) webhost and just paying for the domain name, e.g. Cong. Pablo John Garcia's onion-skinned.com

The advantages are many. There is the aspect of security: you can be assured that your data is protected by the hardware Google, Inc. owns. Google Search also loves content on Blogspot. (I am not saying that there is preferential treatment for material found at Blogspot, but experience showed me that if you want something to be indexed quickly by Google, post it on a Blogger.com-hosted site and not elsewhere. Perhaps this has something to do with the fact that the Google bot and the Blogger files live, if not on the same cluster of servers, at least in the same physical location. Am I making any sense here? :))

EDIT: You can even have a front page for the blog.

Like this post? Please help me pay for my Internet access.
Links to this post
Labels: blogger.com

It seems that for PBwiki you can hide not only its SideBar or the edit link in SideBar, but the search box as well. (This is useful if you are using Google CSE for your PBwiki site.)

The hide the search box, use the following code: <style>.search-box { display:none; !important } <style>

I think this can be done for the other divs too. The process remains the same: use the CSS !important directive, and identify the target div to be hidden.

Like this post? Please help me pay for my Internet access.
Links to this post
Labels: pbwiki, search bar

Go to http://yourwiki.pbwiki.com/AllPages and choose Delete. You can't delete the FrontPage.

Like this post? Please help me pay for my Internet access.
Links to this post
Labels: page delete, pbwiki

Upon creation of a new (pre-2.0) PBwiki, your QuickBar looks like this:


You can't just delete (how?) QuickBar and recreate it with your new content or change its content right away in order to change this QuickBar look. Instead, do the three things given (edit front page, create a new page, and invite another person) and you will be presented with a "hide message" link after you've completed all three after which your code for QuickBar will be reflected. (To invite go to http://yourwiki.pbwiki.com/share.php - the email address you enter for those invited could be completely bogus and you will complete the three todo's. EDIT: Sometimes, this last step does not actually occur. I'm wondering what could be causing this. I just resorted to deleting the QuickBar itself.)

Like this post? Please help me pay for my Internet access.
Links to this post
Labels: pbwiki, quickbar problem

If you ever needed to identify the color on screen in one-click, try Instant-Eyedropper.

You can also use ColorPic (reviewed by Throx here).

Like this post? Please help me pay for my Internet access.
Links to this post
Labels: color picker, colorpic, instant eyedropper, onscreen color

I noticed that my post How to track your Blogger statistics using Google Analytics was deemed older by the system than This blog is getting popular :-) even though the latter was posted before the former. I think the reason for this was that I started working on the former first. So for Blogger.com, what really matters is the first draft that it had in the system. (I could be mistaken though.)

Like this post? Please help me pay for my Internet access.
Links to this post
Labels: blogger posts

This blog had been receiving a lot of traffic the past few days. I based this on the Feedjit Live Traffic Feed, which sometimes show me a completely different set of last 10 visitors within a span of four hours. Taking the conservative side, that would indicate at least 30 unique visitors every day, not bad for a site which is just a collection of my tech notes :) (But actually how much traffic? I don't know, which is why I am installing Google Analytics on this blog.)

The number 1 referrer is Google Search itself, and most entries referred to are those which I just wrote for myself originally so that I can refer to them in the future, for example, How to send email via MyBB using fsockopen and GMail or How to prevent automated registrations in MyBB.

What is also surprising is that some entries are available from Google Search within hours after being posted. For example, Symfony could not perform XLST transformation was posted this afternoon (Philippine time) after I had problems running the Sandbox application of Symfony; tonight it is already seen via a search from Google.pl

Like this post? Please help me pay for my Internet access.
Links to this post
Labels: popularity of blog, self-promotion

This blog is getting popular, which prompted me enough to install Google Analytics on this blog. As you might have noticed, one of the differences between Blogger and Wordpress.coms is that Wordpress comes with builtin statistics (I love that page) but Blogger does not have something like that - you have to install a third-party plugin. (I had a choice between StatCounter and Google Analytics, but since I'm using Google Analytics for my other sites - PhilTowns.net, EduPhil.org - I reckoned it would be a lot easier to have the same account manage different sites.)

To install the Google Analytics code, all you have to do is to go your Edit Layout page and look for an HTML/JavaScript page element. Then copy the code from the Analytics site, and that's it.

Like this post? Please help me pay for my Internet access.
Links to this post
Labels: blogger.com, google analytics

And keeps an alert on "pbwiki" (or perhaps "pbwiki.com"). At least that's how I would explain the presence of an entry in my Feedjit Live Traffic Feed which says it came from http://mail.google.com/a/pbwiki.com

I know I failed to get a snapshot of that, but since this is going to be indexed by Google, someone from PBwiki will be coming here and will register in the Feedjit Widget again :) By that time I'll make sure I will be able to take a snapshot.

Like this post? Please help me pay for my Internet access.
Links to this post
Labels: feedjit, google alerts, pbwiki

One can consider a module as a set of actions which can be done on a database table. For a weblog application for example, there would be a comment table (containing the comments in its rows) and a post table (containing the posts in its rows). One would expect then that there would be a comment module and a post module.

Modules which act on related tables could then be considered as an application.

Like this post? Please help me pay for my Internet access.
Links to this post
Labels: module vs application, symfony

The beauty of Symfony as a web application framework is the separation of the logical code (the one which does the business decisions) from the presentation code.

Logical code can be found in a module's actions folder, while the presentation code can be found in the module's templates folder.

Like this post? Please help me pay for my Internet access.
Links to this post
Labels: logic and presentation, symfony

There are two kinds of code generation available in Symfony: one for scaffolding, and another for administration.

A scaffolding is the basic structure (that is, the actions and templates) required to run a CRUD on a table. Scaffolding code is minimal and should be seen as a starting base: you should only use it during the development phase of the application so you have web access to the database, or build a prototype.

An administration (specifically backend administration) is a sophisticated interface for data manipulation. Presentation is emphasized for administration code (vis-a-vis scaffolding code). Sorting, pagination, filtering, and other features are typically utilized in administration interface.

Scaffolding code is for use by the developer; administration code is targeted at the end-user. Scaffolding code is meant to be modified manually; administration code is changed via configuration and inheritance.

Like this post? Please help me pay for my Internet access.
Links to this post
Labels: administration, scaffolding, symfony code generation

The meta headers for a Symfony project can be found in /apps/frontend/config/view.yml

From this file, you can change the content-type, title, instruction for robots, description, keywords, and language.

Like this post? Please help me pay for my Internet access.
Links to this post
Labels: symfony

Edit /apps/frontend/config/routing.yml and change the homepage rule as follows:

homepage:
url: /
param: { module:

, action: index }

Like this post? Please help me pay for my Internet access.
Links to this post
Labels: symfony

I had a problem with spam registrations at the Education in the Philippines Forum (http://eduphil.org/) for the last two weeks. I am enabling CAPTCHA images for registration by logging in as administrator and going to Board Settings > Change > General configuration and looking for the setting "CAPTCHA Images for Registration & Posting".

Like this post? Please help me pay for my Internet access.
Links to this post
Labels: automated registration, mybb

You can edit the default MyBB registration agreement by logging in as administrator, and going to Language Packs > Manage. Then for Editing Options click on Go. Then click on the edit link for member.lang.php. The title is agreement, and the entire registration agreement is broken up into agreement1 to agreement5.

Like this post? Please help me pay for my Internet access.
Links to this post
Labels: customizing mybb

1. To initialize the data model, create a schema.yml file in project_dir/config/. This schema.yml file is in YAML format. Then build the model using "php symfony propel-build-model" (without the quotes). To create the SQL queries, use "php symfony propel-build-sql", and to insert those queries into the actual database, use "php symfony propel-insert-sql". project_dir/config/databases.yml contains the information as to kind of database.
2. Symfony can generate CRUD scaffolding: "php symfony propel-generate-crud ", e.g. "php symfony propel-generate-crud frontend post Post".
3. To clear the cache, you can use "php symfony cc" or "php symfony clear-cache".

Note: Make sure you are in the project's root when using the symfony command.

Like this post? Please help me pay for my Internet access.
Links to this post
Labels: symfony

In a previous post I described how to install Symfony in a WAMP environment. Because the workstation that I am using right now had been my workstation for years, and had seen a lot of abuse, there's bound to be an error regarding configuration files.

I was following Symfony-Project.org's one-hour tutorial for one's first Symfony application when I received the following message after I issued php symfony propel-build-model:

[propel-om] Could not perform XLST transformation. Make sure PHP has been compiled/configured to support XSLT.

The next step (php symfony propel-build-sql) would lead to an error, with the system not finding the required SQL file supposed to be output by the previous one.

I did a php -m and indeed, xsl was not one of those listed. To know which php.ini file the system is using, I issued php --ini. It indicated that my config file is C:\wamp\bin\php\php5.2.5\php.ini. So I opened this file, uncommented the line extension=php_xsl.dll, and to be sure, copied this INI file to my Windows directory.

Like this post? Please help me pay for my Internet access.
Links to this post
Labels: php xslt

Symfony is a robust PHP applications framework. To install it on Windows XP, I used the following steps:

1. Download the WAMP server from http://www.en.wampserver.com/
2. Install WAMP server - use default config.
3. Test WAMP server - basic phpinfo.php in C:\wamp\www\. Note: in one of the three workstations I followed this procedure, I could not get WAMP to run because there was a prior installation of Apache2Triad.
   
4. Download the Symfony Sandbox application from Symfony-Project.org
   
5. Extract the sandbox application to C:\wamp\www and then navigate to the projects "web" directory. You should see the default "Symfony Project Created" page.
6. Install PEAR on Windows - open a command line window and navigate to C:\wamp\bin\php\php5.2.5 and run the go-pear.bat file. Just follow the prompts. You should be connected to the Internet.
7. pear channel-discover pear.symfony-project.com
8. pear remote-list -c symfony
9. pear install symfony/symfony - last part after number of bytes will take some time
10. pear config-show (?) - add C:\wamp\bin\php\php5.2.5 to system path (My Computer > Properties, then open a new CLW)
11. cd c:\wamp\www
12. mkdir myproject
13. cd myproject
14. symfony init-project myproject
15. copy C:\wamp\bin\php\php5.2.5\PEAR\data\symfony to c:\wamp\www\myproject\web\sf - seems to be unnecessary?
16. symfony init-app myapp - at this time you can already go to http://localhost/www/myproject/ which will give a default page
17. symfony init-module myapp mymodule

Like this post? Please help me pay for my Internet access.
Links to this post
Labels: symfony, Windows

So I spent more than two hours trying to make a design work, only to realize after I achieved the required form that it is pretty close to useless since the output won't be like it 90% of the time. Moral lesson: put the end in sight.

Like this post? Please help me pay for my Internet access.
Links to this post
Labels: realizations

To remove the edit link in PBwiki's sidebar, put this in the sidebar itself: <style>.sb_note { display:none; !important } </style>. The main difference with this from the technique in hiding the sidebar itself is that you really need to put the !important CSS declaration. Again, this was not tested on PBwiki 2.0. You can still edit your sidebar using http://yourwiki.pbwiki.com/SideBar?edit=1

Like this post? Please help me pay for my Internet access.
Links to this post
Labels: customizing pbwiki

If you have three tabs in your sidebar, then you have version 2. Put this in the pages where you don't want the sidebar to appear:
<style>.sidebar_v2 { display: none; } </style>

Otherwise, use this: <style>#SideBar { display:none; } </style>.

If you want to hide the sidebar for all pages, then put the necessary code in the sidebar itself. You can still access the sidebar page via http://yourwiki.pbwiki.com/SideBar

P.S. This was not tested for PBwiki 2.0.

Like this post? Please help me pay for my Internet access.
Links to this post
Labels: customizing pbwiki, pbwiki, sidebar

A few moments ago, I posted a review of Contactify.com, which offers free contact forms without exposing your email address. The default Contactify.com form is hosted at their site, but you can embed the widget in your own site.

Well, I have tried installing it here at Isles Tech, and this is what I got:

Take note of that contact form on the right side. I thought there's an easy way to change the default settings (e.g. width, color, etc.) but apparently there's none; the Contactify.com website states that you need to change the CSS file - apparently I don't have time for this :). So for the time being, I will just be happy putting a link on my right sidebar to my contact form at the Contactify site itself.

This reminds me of Google's hardcoding of the iframe height in their Custom Search Engine.

Like this post? Please help me pay for my Internet access.
Links to this post
Labels: design notes

I was browsing through some stuff related to a market research for my latest possible venture (I am doing a thorough study this time, unlike before) when I chanced upon Contactify.com

The idea is innovative: you are provided with a private contact form, which people can use to contact you without revealing your email address. (With the amount of spam we are receiving nowadays, I think no one who really knows what is happening is willing to put his email address in the open.) So instead of giving your email to an audience, you can just give your Contactify link; mine is http://www.contactify.com/751f1 (yep, that's live). Also, once you get tired of a given link (or when it is exploited also by spammers - although it is more difficult it is still possible), you can create another link.

When someone clicks on your Contactify link, they are brought to a page on the Contactify site where they can find a form to contact you. The form has captcha security, so bots are kept out. The messages posted on your Contactify contact form are sent to your email.

You can even embed the Contactify form on your site; for a live example, see BrianHall.com's contact form. (I will be adding the Contactify widget on this site after I post this entry, so you might have already seen that form on the side.)

I did notice though that the captcha system of the site seem to have some problems; my test message to myself took me six tries to get right, and I could not have had suddenly develop a bad eye sight at that time.

See also

• KillerStartups.com's review of Contactify.com
• SiteGuide.us' review
• Online Tech Tips' review
  

Like this post? Please help me pay for my Internet access.
Links to this post
Labels: contactify.com, email without address, free contact form

To add meta tags (such as keyword and description) to MediaWiki pages, for all pages, you have to edit monobook.php which is located in the skins directory. Look for the part which says <title><?php $this->text('pagetitle') ?><title> and add the meta tags after that. You can also use this area to add other meta tags (such as the verify tag for Google Webmasters Tools) and tracking scripts (such as Google Analytics or StatCounter).

If you want to have a more flexible solution, which can work on individual pages, you can use Supernifty.com's MediaWiki meta extension, which is based on Jim R. Wilson's meta keywords extension. The instructions on installation and use can be found on the Supernifty page.

Like this post? Please help me pay for my Internet access.
Links to this post
Labels: MediaWiki, meta tags

ClickToJoin.com, BeAHero.com

Like this post? Please help me pay for my Internet access.
Links to this post
Labels: domain name

MyBB mod for creating Google sitemaps

Note: this must be installed on the same level as your index.php and forumdisplay.php files.

Like this post? Please help me pay for my Internet access.
Links to this post
Labels: mybb sitemaps

You can use the Dynamic Meta mod for MyBB to add keyword and description meta tags. (how to install MyBB mods)

If you have to add other meta tags (e.g. for Google Webmaster Tools), you can edit the headerinclude template: Admin CP > Templates > Modify/Delete > Expand > headerinclude.

Like this post? Please help me pay for my Internet access.
Links to this post
Labels: mybb

To install a MyBB mod, you must extract the package and upload it to your inc/plugins directory.

Then go to your Admin CP > Board Settings > Plugin Manager and activate the appropriate plugin.

You must go to Board Settings > Change to change the settings for the plugin.

Like this post? Please help me pay for my Internet access.
Links to this post
Labels: mybb