Ben Langhinrichs

Photograph of Ben Langhinrichs

E-mail address - Ben Langhinrichs







Recent posts

Sat 18 Jul 2020

Enduring favorite - Getting Data out of Notes (for whatever reason)



Thu 9 Jul 2020

Maximizing power while minimizing code and effort



Fri 29 May 2020

Round tripping, even while staying put


August, 2020
SMTWTFS
      01
02 03 04 05 06 07 08
09 10 11 12 13 14 15
16 17 18 19 20 21 22
23 24 25 26 27 28 29
30 31

Search the weblog





























Genii Weblog

Browser confusion

Tue 24 Feb 2004, 07:57 AM



by Ben Langhinrichs
When developing applications for the web, I am well aware that there are cross platform issues.  There are more issues that I care to think about just between IE 5.5 and IE 6.0, which blows my mind.  Given that, I took a look at visitors to my site in the past 28 days, and the following statistics stand out (these all have at least one hit):
Browser
1.Explorer 6.x69.1 %
2.Explorer 5.x14.6 %
3.Mozilla 1.x9.9 %
4.Netscape 3.x3.1 %
5.Opera 7.x1.4 %
6.Netscape 7.x0.6 %
7.Safari 1.x0.5 %
8.Netscape 4.x0.5 %
9.Opera 6.x0.1 %
10.AvantGo 5.x0.0 %
11.Konqueror 3.x0.0 %
12.Lotus-Notes 5.x0.0 %
13.Mozilla 0.x0.0 %
14.Lotus-Notes 6.x0.0 %
15.MS FrontPage 4.x0.0 %
16.BorderManager 3.x0.0 %
17.Netscape 6.x0.0 %
18.iCab 2.x0.0 %
19.Windows 9.x0.0 %
20.Unknown0.0 %
21.WebTrafficExpress 1.x0.0 %


Now, I don't even know which browsers are which, in some cases.  That Lotus-Notes 6.x I can identify, but Windows 9.x? MS FrontPage 4.x has its own browser?  Netscape 3.x/4.x/6.x/7.x?  iCab 2.x?So, which of these are brand new cutting edge browsers, and which are antiquated, time to upgrade browsers?  Which support CSS in all its glory, and which are lucky to support tables?  It would be easy to dismiss all but the top three, but what is the "Firefox" everyone is talking about?  Is that Mozilla 1.x or Netscape 3.x?

Beyond what I see on my site, which of these do I need to test to responsibly declare that an application is cross browser compatible?  Do I have to test just the top two, three, four?  Is there a simpler chart that says these fall into four or five basic categories using the same engines?  Why is it all so complicated out there in the world of standards?

Copyright 2004 Genii Software Ltd.

What has been said:


119.1. Scott
(02/24/2004 06:53 AM)

You have to remember that the user agent string can be set by the client, depending on the software they're using to get to your site. A web crawler will have something set as its agent...a good one will have something descriptive, some do not. If memory serves me, Opera can let you set whatever you want as its useragent.

FrontPage does show its own useragent string when it pulls a page.


119.2. Ben Langhinrichs
(02/24/2004 07:13 AM)

So, BorderManager, WebTrafficExpress and those sorts are just set by the user agent. That cuts it down by a good bit.


119.3. Simon
(02/24/2004 02:36 PM)

FireFox (.8)/Firebird(.6)/Phoenix(.5) is a Browser only based on Mozilla 1.x. In next Mozilla increment is suppossed to replace the browser.


119.4. Rock
(02/25/2004 08:10 AM)

You also have to remember that your numbers are skewed based on your own hits to your site. So, for me the numbers break down as follows:

MS Internet Explorer63.9 %

Mozilla 23 %

Unknown 9 %

Safari 41.3 %

Opera 1.1 %

Netscape 0.9 %

Lotus Notes web client 0.1 %

Now a part of that is me hitting my own site; a part of that is that 50% of my traffic is non-US, and non-US readers tend to not be so heavily IE-centric. So, my point is that the IE traffic you're seeing is probably strong, but not as strong as the numbers appear.

Rock


119.5. Rock
(02/25/2004 08:10 AM)

Ooops, Safari should be 1.3%, not 41.3%. Sorry. :-(

Rock


119.6. Ben Langhinrichs
(02/25/2004 02:31 PM)

Heck, Rocky, I was already to be very impressed by your greater than 100% readership. It must be the gay marriage post.

As to your point, this is an insignificant factor for me, as I barely ever go to my own blog through a browser (remember, I have my own design and create everything in Notes). Even if I did, I only use IE 5.5, so the results for IE 6.x wouldn't be skewed.