http://www.spreadfirefox.com/en-US/worldrecord

Download Day is here!
Set a Guinness World Record
Enjoy a Better Web

Sounds like a good deal, right? All you have to do to help us set the record for the most software downloaded in 24 hours is get Firefox 3 now – it’s that easy. We’re not asking you to swallow a sword or to balance 30 spoons on your face, although that would be kind of awesome.

Please download Firefox 3 by 17:00 UTC on June 18, 2008. That’s 10:00 a.m. in Mountain View, 1:00 p.m. in Toronto, 2:00 p.m. in Rio de Janeiro, 7:00 p.m. in Paris, Madrid, Berlin, Rome and Warsaw, 9:00 p.m. in Moscow, and June 19, 2008 at 1:00 a.m. in Beijing and 2:00 a.m. in Tokyo.

Now I want to share with youthis little quick start guide and won’t tell you what SVN is and it’s whole description.
I have a PHP project in /var/www/html/docs/haibuy. And Apache has Virtualhost haibuy. My IP in LAN is 192.168.1.7.

So now we need to create ‘repository’ where our projects will ‘live’. We can create just one repository for all projects. So I’ve created /svn directory. And then run a command:

svnadmin create /svn/repository

And got repository subdirectory in /svn with its subdirectories structer.

Now we need to import a project to our repository. To use all the features of SVN including branches, merging branches we should create our project directories structure like this:

/path/to/project
          |- trunk
          |- branches
          |- tags

trunk must have you project files.
Now we need to import our project to repository:

svn import /path/to/project file:///path/to/repos/project -m "Comment"

In my case it was:

svn import /var/www/html/docs/haibuy file:///svn/repository/haibuy -m "Comment: initial import"

Now we can delete our project from /var/www/html/docs/haibuy (or if you’re afraid just rename it).
To work with SVN as client I use Eclipse with Subclipse plugin.
I also use PHP nightly plugin for it. To install them both run Eclipse. Then go to “Help->Software Apdates->Find and install”, select “Search for new freatures to install” add “New remote site…” “PHPEclipse Nightly” with URL:
http://update.phpeclipse.net/update/nightly and “Subclipse” with URL:
http://subclipse.tigris.org/update_1.2.x. After that Subclipse and PHP Eclipse will be installed. Reopen your Eclipse and go to Window->Open Perspective->Other… select PHP. Now you have nice PHP environment. Let’s create new project ‘haibuy’: New->Project->SVN->Checkout Projects from SVN. Create a new repository location. As our repository is on the same machine we can add it using such repository location: file:///svn/repository/haibuy/trunk. Then “Next”, select the top folder againt “Next”, and select where to save your work copy. I typed in: /var/www/html/docs/haibuy, so my working copy is on “web-server”.
Now you can use Team submenu in context menu (right mouse click on project or project files) to work with SVN. To get the latest version from repository use Team->Update. To confirm your changes and upload them to the repository use Team->Commit…

Now we need to make our repository reachable from other machines. We need to open Firewall and add 3690 port TCP protocol in ‘Other Ports’ section. Also we can check 80 port in ‘Trusted Services’ to make our http://haibuy available over the LAN.
After that we have to start demon, which will listen to the 3690 port and allow to work remotly with SVN.

svnserve -d

Now you can work with repository from other machines. path in my case: svn://192.168.1.7/svn/repository/haibuy/trunk

Other developer is working under Windows XP and he uses TortoiseVPN client. Everything works fine.

P.S.: to check if your 3690 port is open and svnserve demon started run shell command “telnet 192.168.1.7 3690″ from other machine.

One ofthe two websites I made long ago, using only HTML, I updated it few times and it was my copyright text with e-mail address in the bottom of the page.
That’s it: www.legenda.com.ua
As you see now, my copyright is deleted, now there is:
Webmaster: Hodan B.I. , Site support
But this “webmaster” is lazy or non-competent, cuz when we look at the source we see in <head>…</head> :
<meta name=”author” content=”Osipov Vladimir”>

Second website was developed by the “Olvia” webstudio, where I used to work.
It is located here www.migovo.com.ua.
When we created it there was link to Olvia’s website, but now we see “B y M a x w e l l V a z o v s k y” in the footer. Who is this one? - hell knows. Maximum work he did was changing some texts - and I doubt it gives him rights to tell that this website was created by him. And he was also lazy or just didn’t know, that in <head>…</head> is
<meta name=”author” content=”Olvia Group”>
and no Maxwells mentioned there.

I’ve mentioned this when recently one friend of mine asked me to have a look at these websites and prepare the commercial proposal for SEO service. When I looked at “old” well known for me pages I was embarassed by impudence and disagreed to co-work with them again. Of course it can be that the bosses don’t know about this, perhaps they just asked these “Maxwells” to change something on websites, and these “Maxwells” decided that now they have rights to call themselves “The Creators”. Well if they do this to someone, then someone will do this to them… sooner or later…

2. Passing "\n" through

In last function we used previousSibling - this is the reference value to the previous neighbor node, you can also use nextSibling to go to the next neighbor node. Actually in DOM printing function we could make another cycle: 'while' instead of 'for':

Penalty lottery was really nervous, but I knew Anelka won't score, better they'd let Shevchenko play.

Photo by Reuters

You need to know that every node has the property nodeType. If it's an element nodeType=1, it it's a text node (and can not have child nodes) then nodeType=3

Now try this code:

Look the output of the function in different browsers and you will see, that IE and Opera DOM tree is like:
[TABLE]
[TBODY]
[tr]
[TD]
[text:sample 1]
[/TD]
[TD]
[text:sample text here...]
[/TD][/tr]
[tr]
[TD]
[text:sample 2]
[/TD]
[TD]
[text:sample text here...]
[/TD][/tr]

but in Firefox:
[tr]
[TD]
[text:sample 1]
[/TD]
[TD]
[text:sample text here...]
[/TD][/tr]
[text:\n]
[tr]
[TD]
[text:sample 2]
[/TD]
[TD]
[text:sample text here...]
[/TD][/tr]
[text:\n]

We have additional text nodes (\n - pagebreaks) after each <tr>. That's why when we try to delete the last child of tbody in IE we delete the tr element but Firefox we delete text node with "\n" and the second click deletes tr element. Also when we add new row we don't use pagebreaks, that's why it's deleting in a Firefox in one click.

If we'll change the table code to:

then we won't have such problem. Other way to fix a little bit our delete function. How to make this? Tomorrow you will know

We will try to write JavaScirpt code to add/remove rows to table but without using DHTML functions (addRow, addCell etc.), we will use functions to manipulate with DOM tree like appendChild, createElement, createTextNode.

so, we have such HTML code:

We want to create a function which will add a row (<tr> ... </tr>) to the end of our table. The only parameter will be the id of TBODY element:

Graphically it will be like:

1. [myTR]
2. [myTD]
3. [myTD] | -> 'sample '+(++i)
4. [myTD2]
5. [myTD2] | -> 'sample text here 2...'
6. [myTR] | -> [myTD]
7. [myTR] | -> [myTD]
              | -> [myTD2]
8. [tid] (tbody) | -> first row of our table
                      | ....
                      | -> [myTR]

Now we can add button, which will run our function:

click on the button and you will see how new rows append to the table
It works fine in all modern browsers (I've tested on FF 1.5, Safari 3, Opera 9, IE 6+)

Now we need also have function to delete the last row.

Add a button to the HTML:

Try :). There is a little surprise you will have in Safari and Firefox. You need double click to delete the row, when in Opera and Internet Explorer it works fine. Also you can see, that rows created using our function AddEndRow are removing well in all browsers. Why is it so? Tomorrow I'll tell you...
to be continued

So, now I'm waiting for Yahoo! answer and really hope things will be fine for Yahoo!

BTW, our www.themetallist.com reached it record visitors count - 2041!
Mainly thanks to the SATYRICON report from Moscow:

Some pix:

Other pictures

Video for "Bespechnyy Angel"

Dylan, Engineer at Google called me yesterday evening and we had 45 minutes conversation. First he asked me to describe my usual working day.

After I was aksed about 2 different situations, we were discussing them. I told what will I do on server side and client side. So the situations were: double click on form button, especially if it's the form of ordering something using credit card. Another one - checking if there is already such username in base by registration, using iframe instead of JSON.

Dylan also asked my about JavaScript closures. But unfortunately I had no idea about that, but read some information after and it's really interesting, I will write about this soon

After that I had a chance to give my questions.

PS. my Ubuntu installation went not well, had problems with GRUB, will try again soon