RTL Web Browser Survey
Todd Lehman
I'm very interested to learn how RTLers see web pages. There's a quick
"RTL Browser Survey" up at
http://www.fibblesnork.com/rtl/survey/
and there are 9 questions. Hopefully the results from this survey will
help web page creators as they make design decisions for their sites.
(All the questions are shown below, but you need to go to the web page
to actually take the survey.)
--Todd
=============================================================================
1. What platform do you most often use for web browsing?
o Macintosh
o PC with a flavor of Microsoft Windows
o PC with a flavor of Unix
o PC with other operating system
o Workstation with a flavor of Unix
o (other)
2. What browser do you use most often?
o Netscape Navigator
o Microsoft Internet Explorer
o (other)
3. How fast is your Internet connection on the machine you most often use
for web browsing?
o Slower than 14.4 Kbps modem
o 14.4 Kbps modem
o 28.8 Kbps modem
o Faster than 28.8 Kbps modem (ISDN, T1, T3, etc.)
4. What screen resolution do you most often use for web browsing? (Choose
the one which most closely matches your setup.)
o Graphics 640 x 480 pixels
o Graphics 800 x 600 pixels
o Graphics 1024 x 768 pixels
o Graphics 1280 x 1024 pixels
o Graphics 1600 x 1200 pixels
o Text-based browser such as Lynx
5. How many colors do you see when you browse the web? (Choose the one
which most closely matches your setup.)
o 2 colors (1-bit)
o 4 colors (2-bit)
o 8 colors (3-bit)
o 16 colors (4-bit)
o 256 colors (8-bit)
o 32,768 colors (15- or 16-bit)
o 16,777,216 colors (24-bit) or higher
o Text-based browser such as Lynx
6. Which most closely describes how you feel about the default browser
window size?
o I always use my browser's default window size.
o My browser's default window size is usually adequate, but I sometimes
have to resize it to make it bigger.
o My browser's default window size is seldom adequate, and I often have
to resize it to make it bigger.
o My browser's default window size is ridiculously small, and I never
use the default window size.
7. As you surf from site to site, do you typically resize your browser
window to best fit the site you are visiting?
o Yes, I think this is just a natural part of web surfing.
o Yes, but I wish I didn't have to resize my window so often.
o No, because I don't like resizing my window.
o No, because my screen resolution is so small that I have to run with
my browser window full-screen all the time.
o (I don't browse the web in graphics mode.)
8. How do you feel about the use of frames in web site design?
o I don't like frames because they're a waste of screen space.
o I don't like frames because they make me scroll too much.
o I don't like frames because they make browsing more confusing.
o I don't like frames because when I drop a bookmark and return to it,
my browser doesn't remember the exact place properly.
o I do like frames when they're used properly.
9. Which most closely describes how you view inline GIF and JPEG images?
(Inline images are meant to appear directly on the web page itself, right
next to the text, such as thumbnails and icons and logos. Inline images are
not those which are linked to as a separate hypertext element, such as a
larger views of a thumbnail, or an image library with text links.)
o They show up automatically on the web pages I view.
o They show up on the web pages I view, but only when I ask for them
explicitly.
o My browser cannot display inline GIF and JPEG images, but it can
launch an external program to view them.
o I can't view GIF or JPEG images from my browser.
[end]
Lou Zucaro
I think something might be set wrong in your script. The button
that's on the page simply launches the default mail program and opens
a blank (new) message window.
leh...@visi.com (Todd Lehman) wrote:
>I'm very interested to learn how RTLers see web pages. There's a quick
>"RTL Browser Survey" up at
> http://www.fibblesnork.com/rtl/survey/
Lou Zucaro
http://www.pause.com | -==[UDIC]==-
|
"I...I think he was a mandroid." |* Lego Dragon *
LEGO: FS+(1819) TO+++1(6399) BO+++(709) TR+++(4558) PA+(6416)
SP+++2(497) CA+++(6071) PI+++c(6276) AQ+++3c(6195) TC++(8857)
#+++++ S++ LS+ Hs M+ YB67m
Todd Lehman
In article <55r4la$5...@Nntp1.mcs.net>, Lou Zucaro <l...@pause.com> writes:
> I think something might be set wrong in your script. The button
> that's on the page simply launches the default mail program and opens
> a blank (new) message window.
Actually, there's no script. The HTML form uses ACTION="mailto:" which
causes your browser to submit the form results via e-mail instead
of to a CGI engine.
So it's either a problem with the way you have your e-mail gateway
configured with your browser, or your browser doesn't support forms
they way I implemented it here. (Or, something else that I don't
understand.)
But it seems to be working fine for other people...got about a dozen
responses so far today.
--Todd
Patrick Delahanty
Todd Lehman wrote:
...
> 1. What platform do you most often use for web browsing?
> 2. What browser do you use most often?
If your ISP has Agent logs, you already have this information...and more
accurate than any survey can tell you. My server logs the agent
(browser) type and platform along with the other information.
Ask your ISP if this information can be made available to you.
> 8. How do you feel about the use of frames in web site design?
...
> o I do like frames when they're used properly.
The problem here is that a lot of people who do frames on their pages
THINK they're doing them properly...but aren't.
Also, I've noticed that there are differences in the way MSIE and
Netscrape display frames. You've got to be careful. ...plus the fact
that Netscrape 2.0 and 3.0 treat the back button differently in frames.
Ethan A Merritt
In article <55qiud$i...@darla.visi.com>, Todd Lehman <leh...@visi.com> wrote:
>
>I'm very interested to learn how RTLers see web pages. There's a quick
>"RTL Browser Survey" up at
>
> http://www.fibblesnork.com/rtl/survey/
>
>to actually take the survey.)
(yeah, I read the above. But where do I put meta-comments on your form?)
>=============================================================================
>8. How do you feel about the use of frames in web site design?
>
> o I don't like frames because they're a waste of screen space.
> o I don't like frames because they make me scroll too much.
> o I don't like frames because they make browsing more confusing.
> o I don't like frames because when I drop a bookmark and return to it,
> my browser doesn't remember the exact place properly.
All of the above. In spades. I HATE frames.
And besides you left out the killer - even if you can see them on the
screen you can't print them because the NetScape print command freaks
when it sees a frame. Maybe they'll fix it later (heck maybe they
already have on some versions, but not the ones I can use).
And Mosaic doesn't like frames either. And Micro$oft whatever it is
doesn't run on my workstations.
>9. Which most closely describes how you view inline GIF and JPEG images?
> o They show up automatically on the web pages I view.
They show up automatically, but usually with an incorrect gamma value
because no-one on the net can decide just what the standard gamma should
be. Doesn't matter for many things, but those cool 24-bit plane JPEG
images of Legoland are wasted if they show up as if they were taken
20 minutes after sundown.
So I end up having to trigger an external viewer anyway so I can
play with gamma.
Ethan A Merritt
mer...@u.washington.edu
(well, you asked :-)
Todd Lehman
Patrick Delahanty <patr...@internet.com> writes:
> If your ISP has Agent logs, you already have this information...and more
> accurate than any survey can tell you. My server logs the agent
> (browser) type and platform along with the other information.
>
> Ask your ISP if this information can be made available to you.
Yup, all that information is available from my server. But in order to
see it, I need to use a CGI script to process the form, and I'm not
really CGI-savvy yet.
But I'm not sure I'd want to use that info anyway, from what I've seen
of it. HTTP_USER_AGENT comes as a line in the e-mail generated by the
ACTION="mailto:" directive, so I've got a collection right now of 162
unique agent strings from various browsers. Some of them are pretty icky!
:-{
Netscape (Mozilla) is pretty good about identifying things accurately but
even there I don't think I'd have much fun trying to figure out what
platform each new and weird browser is running on each time a new one
showed up, and then maintain a database file mapping the string into a
platform and OS. Some samples:
Mozilla 2.0 (X11; I; HP-UX A.09.03 9000/715)
Mozilla 2.0 (X11; I; IRIX 5.3 IP12)
Mozilla 2.0 (X11; I; OSF1 V3.2 alpha)
Mozilla 3.0 (X11; I; Linux 1.3.48 i586)
Mozilla 2.0 (X11; I; IRIX 5.3 IP22)
Mozilla 2.0 (X11; I; SunOS 5.4 sun4m)
Mozilla 2.0 (Macintosh; I; PPC)
Mozilla 2.0 (Macintosh; I; 68K)
Mozilla 3.0 (Win95; I)
Mozilla 2.0 (Win16; I)
IBrowse 0.70 (Amiga; Always 32-bit)
AIR Mosaic (16-bit) version 4.16.09.32
Prodigy Internet GW(v0.9beta) - ae01dm04sc03
I also have to keep close tabs on processors this way -- for instance
knowing that Linux on a "i586" probably means Linux on a Pentium, which
probably means a PC compatible. I just don't want to have to worry
about all that...too much work. I guess the script could ask the user
"Hey, I don't recognize your browser identity string, could you tell me
some things about your platform?" but I think that's more complicated
than I want to get.
I just figured asking explicitly was the best way, and that people
hopefully wouldn't mind answering that question.
>> 8. How do you feel about the use of frames in web site design?
>> o I do like frames when they're used properly.
>
> The problem here is that a lot of people who do frames on their pages
> THINK they're doing them properly...but aren't.
>
> Also, I've noticed that there are differences in the way MSIE and
> Netscrape display frames. You've got to be careful. ...plus the fact
> that Netscrape 2.0 and 3.0 treat the back button differently in frames.
Hmm, good point. Yup. What are some do's and dont's of frame design
for LEGO pages?
I guess one really good "do" of frame design is to allow the user to
disable frames, as Joshua has done in his reviews site. Waycool.
--Todd
edb...@aol.com
In article <55qiud$i...@darla.visi.com>, leh...@visi.com (Todd Lehman)
writes:
>1. What platform do you most often use for web browsing?
> o PC with a flavor of Microsoft Windows
>2. What browser do you use most often?
>
> o Netscape Navigator
>
>3. How fast is your Internet connection on the machine you most often
use
> for web browsing?
> o 28.8 Kbps modem
>4. What screen resolution do you most often use for web browsing?
(Choose
> the one which most closely matches your setup.)
> o Graphics 800 x 600 pixels
>5. How many colors do you see when you browse the web? (Choose the one
> which most closely matches your setup.)
>
> o 32,768 colors (15- or 16-bit)
>6. Which most closely describes how you feel about the default browser
> window size?
> o My browser's default window size is seldom adequate, and I often
have
> to resize it to make it bigger.
>
>7. As you surf from site to site, do you typically resize your browser
> window to best fit the site you are visiting?
>
> o Yes, I think this is just a natural part of web surfing.
>
>8. How do you feel about the use of frames in web site design?
>
> o I do like frames when they're used properly.
>>
>9. Which most closely describes how you view inline GIF and JPEG images?
>
> o They show up automatically on the web pages I view.
Ed "Boxer" Jones
The quickest way to a man's heart is through his chest - Roseanne
Check out my castle at: www.is.co.za/foxx
Patrick Delahanty
Todd Lehman wrote:
> Patrick Delahanty <patr...@internet.com> writes:
> > If your ISP has Agent logs, you already have this information...and more
> > accurate than any survey can tell you. My server logs the agent
> > (browser) type and platform along with the other information.
> >
> > Ask your ISP if this information can be made available to you.
>
> Yup, all that information is available from my server. But in order to
> see it, I need to use a CGI script to process the form, and I'm not
> really CGI-savvy yet.
>
> But I'm not sure I'd want to use that info anyway, from what I've seen
> of it. HTTP_USER_AGENT comes as a line in the e-mail generated by the
> ACTION="mailto:" directive, so I've got a collection right now of 162
> unique agent strings from various browsers. Some of them are pretty icky!
> :-{
You misunderstood.
You don't need to know CGI at all! Just look at the server's log file.
If it's done correctly, there should be a USER-AGENT field with the type
of browser listed. (...and yes, the information isn't as plain as "I
use Netscape 3.0 and I use a Mac." ...but the information is there and
accurate.)
So essentially, you can get that info by looking at the log file and
don't have to write a CGI to collect that data. ...and with the log
file, you'd have stats on EVERYONE...not just people who filled out a
form.
My log file on sebago.internet.com collects that data. Every so often I
run a stat program that counts up uses of particular browsers. Ask
Suzanne to show you her stats.
(Also, some browsers (including certain versions of Netscape) have bugs
in the ACTION="mailto:..." where it won't work...just thought I should
warn you. Internet Movie Database had trouble with that a while back.)
> Hmm, good point. Yup. What are some do's and dont's of frame design
> for LEGO pages?
The first thing I usually tell people who ask about frames is "Don't do
frames just for the sake of doing frames." I've seen lots of sites with
frames that don't really need them. Sure, they might make a site look
really cool, but in the end, they confuse the user, slow down the site,
and don't really help much.
Having a toolbar at the bottom of a page is just as good as, and often
better than, having one in a frame.
Don't take away scrollbars. If someone has a smaller screen, they may
need them. (For example, I can't even read the stuff on the right of
the top of your Fibblesnork guide. I don't like to resize my window to
the full screen since I usually have other windows open too.)
Also, web sites with frames should check EVERY link and make sure they
open in the right frame...or open in the complete window. I've seen
some pages out there where people say "Click here for a cool site" and
it opens someone else's page within that frame. They obviously forgot
to use TARGET="_top". Most annoying.
Over use of frames is also a problem on a lot of sites. I've seen a few
where there are a lot of little frames created in a window. This may
have looked great on someone's screen at one point...but on smaller
screens, everything gets mashed together.
Also, in one frame, it's good to have a menu for people to get back to
the previous page. In Netscape 2, they're not able to use the back
button to go back in a frame. (People never really caught on to the
pop-up menu with "Back in frame".)
> I guess one really good "do" of frame design is to allow the user to
> disable frames, as Joshua has done in his reviews site. Waycool.
Definately. Don't kick people out if they don't have frames. It's
pretty easy to make a version of ANY web site without frames...and still
maintain pretty much the same design.
I've seen a LOT of sites (including my company) have two different
versions of their sites. One for frames and one for no frames. While
this does allow you to do special things for the frames people...or dumb
stuff down for noframes people, it's not necessary at all. It just
means that you have to change something twice if a page changes.
What I've done with some of my pages (like the "CHiPs" FAQ) is design
them so that people with or without frames end up seeing the same
pages...they're just presented differently. That makes the FAQ a lot
easier for me to maintain.
I could go on and on...
Todd Lehman
Patrick Delahanty <patr...@internet.com> writes:
>
> You misunderstood.
>
> You don't need to know CGI at all! Just look at the server's log file.
> If it's done correctly, there should be a USER-AGENT field with the type
> of browser listed. (...and yes, the information isn't as plain as "I
> use Netscape 3.0 and I use a Mac." ...but the information is there and
> accurate.)
I'm not a big fan of that approach because caching proxy servers throw
stats out of whack in ways that aren't measurable.
> So essentially, you can get that info by looking at the log file and
> don't have to write a CGI to collect that data. ...and with the log
> file, you'd have stats on EVERYONE...not just people who filled out a
> form.
My understanding of httpd logs is that they only record accesses to the
server, not requests from clients. The server can't know about all
requests from clients because there can be any number of caching servers
in-between.
Looking at my most recent server log, 3% of the hits on my site have
come from AOL's proxy servers alone. That gives me kind of a woogly
feeling about using the server log for stats.
The only way I know of to guarantee a connection between a given client
and a target http server is to use CGI on the target server. Are there
other ways? I'm still pretty new to this whole logging biz...
Anyway, it's really not that big of a deal. I think people won't mind
answering the platform and browser questions.
> My log file on sebago.internet.com collects that data. Every so often I
> run a stat program that counts up uses of particular browsers. Ask
> Suzanne to show you her stats.
>
> (Also, some browsers (including certain versions of Netscape) have bugs
> in the ACTION="mailto:..." where it won't work...just thought I should
> warn you. Internet Movie Database had trouble with that a while back.)
Yeh...LouZ had some trouble yesterday. I'll move it to CGI sometime,
but it probably won't be until December...
--Todd
PatrickD
Todd Lehman wrote:
> Patrick Delahanty <patr...@internet.com> writes:
> > You misunderstood.
> >
> > You don't need to know CGI at all! Just look at the server's log file.
> > If it's done correctly, there should be a USER-AGENT field with the type
> > of browser listed.
>
> stats out of whack in ways that aren't measurable.
Actually, the proxies just append a text string to the end and don't
change the browser/platform information. They don't mess that up.
> My understanding of httpd logs is that they only record accesses to the
> server, not requests from clients. The server can't know about all
> requests from clients because there can be any number of caching servers
> in-between.
"any number"? Well, I suppose there could...but it's unusual for there
to be more than one. If you start caching someone else's cache, the
data gets old and out of date fast.
> Looking at my most recent server log, 3% of the hits on my site have
> come from AOL's proxy servers alone. That gives me kind of a woogly
> feeling about using the server log for stats.
Actually, it is accurate. Unless you have a wildly popular site like
CNN, CNet, or Ziff Davis, they won't cache it. Every time someone from
AOL hits any of my sites, it does get added to the log.
> The only way I know of to guarantee a connection between a given client
> and a target http server is to use CGI on the target server. Are there
> other ways? I'm still pretty new to this whole logging biz...
That's the only way to absolutely guarantee a connection, but as I said,
AOL will connect anyway. (...and if hits from AOL end up being off by a
couple, then it's STILL more accurate than any web counter or an e-mail
survey!)
Todd Lehman
Gulp, well this is starting to get above my head and beyond my knowledge
level of this...I'm going to have to bail from this discussion after this
reply... And rec.toys.lego isn't the right forum for this...
ObLEGO: At a second-hand toy store a few months ago, I found a red LEGO
storage bin that looks sort of like a lunch box and sort of like a little
crayon or pencil box for kids. It's got a flip-up lid on the side.
There's no copyright date but it does say "LEGO(R)". Anyone know when
this was made?
PatrickD <patr...@internet.com> writes:
>> I'm not a big fan of that approach because caching proxy servers throw
>> stats out of whack in ways that aren't measurable.
>
> Actually, the proxies just append a text string to the end and don't
> change the browser/platform information. They don't mess that up.
I'm thinking more of *caching* proxy servers than of regular proxy
servers.
But my server logs unfortunately aren't that comprehensive anyway. All
I see is something of this format on each line:
hostname - - [date:time -tz] "GET filename" result bytes-transferred
I do have access to all the information suggested a couple messages
back, but I still need to use CGI to get at it. I don't have control
over how my ISP creates the server log (I'm using a web hotel which I
telnet to, not my own actual server).
>> Looking at my most recent server log, 3% of the hits on my site have
>> come from AOL's proxy servers alone. That gives me kind of a woogly
>> feeling about using the server log for stats.
>
> Actually, it is accurate. Unless you have a wildly popular site like
> CNN, CNet, or Ziff Davis, they won't cache it. Every time someone from
> AOL hits any of my sites, it does get added to the log.
Hmm. When I set up a proxy server through visi.com, I had to explicitly
tell it -not- to cache the stuff on fibblesnork.com because I was working
on the pages. It caches everything by default (they have something like
a 6 gig cache).
>> The only way I know of to guarantee a connection between a given client
>> and a target http server is to use CGI on the target server. Are there
>> other ways? I'm still pretty new to this whole logging biz...
>
> That's the only way to absolutely guarantee a connection, but as I said,
> AOL will connect anyway. (...and if hits from AOL end up being off by a
> couple, then it's STILL more accurate than any web counter or an e-mail
> survey!)
I still think the survey is much nicer than the logs. The logs only
tell me what browser/platform the user happened to be using when they
filled out the form. For all I know they just happened to be using a
non-LEGO-friend's 1024x768 24-bit Mac at work on a T1 one night, when they
normally do 99% of their web browsing from home on a 28.8 modem on a PC
with Win95 and a 640x480 8-bit video card.
Don't get me wrong -- I think the browser/platform logging thing in the
server logs is really neat -- I just don't think it matches up nicely
with the type of survey I want. I don't think it's appropriate here.
I also like to have the platform/browser data precisely aligned with the
other fields, so I can separate out the data sets if I want to -- for
instance to see that, say, 70% of Mac users have 1024x768 screens or
higher, while only 50% of PC users have screens that large. Or, maybe,
that Netscape users tend to like frames better than MSIE users. I don't
see how I would be able to do this sort of analysis if I was gleaning
this info from the server logs.
--Todd
Zach Babayco
In article <56554a$n...@darla.visi.com> leh...@visi.com (Todd Lehman) writes:
>
>ObLEGO: At a second-hand toy store a few months ago, I found a red LEGO
>storage bin that looks sort of like a lunch box and sort of like a little
>crayon or pencil box for kids. It's got a flip-up lid on the side.
>There's no copyright date but it does say "LEGO(R)". Anyone know when
>this was made?
>
textured feel to it, or is it smooth?
--
Zach Babayco
za...@netcom.com <-------finger for PGP public key
http://www.geocities.com/SiliconValley/Park/4127
-----
If you need to know how to set up a mail filter or defend against
emailbombs, send me a message with the words "get helpfile" (without the
" marks) in the SUBJECT: header, *NOT THE BODY OF THE MESSAGE!* I have
several useful FAQs and documents available.
Todd Lehman
Zach Babayco <za...@netcom.com> writes:
> It sounds like one I had around 1984-85 or so. Does the box have a
> textured feel to it, or is it smooth?
Textured everywhere except the background of the LEGO logo.
--Todd
Todd Lehman
I wish I knew what. All I'm doing is METHOD="post" ACTION="mailto:..."
in the HTML. Pretty simple thing...
I know the technique works because I've received 1500+ form responses
over the past few months from the Fibblesnork LEGO Survey, the Cool LEGO
Site of the Week voting, and from this RTL Web Broswer Survey.
(But obviously this stinks compared to CGI.)
> The button
> that's on the page simply launches the default mail program and opens
> a blank (new) message window.
What it's supposed to do is launch the default mail program and send the
form data via mail. If it's not doing that, you might have your browser
mail gateway set up wrong?
I dunno. I wish I could help. Beyond my knowledge here, and I don't
plan on keeping it in this form too much longer. Gonna go CGI at some
point.
--Todd
PatrickD
Todd Lehman wrote:
> Lou Zucaro <l...@pause.com> writes:
> > I think something might be set wrong in your script.
>
> I wish I knew what. All I'm doing is METHOD="post" ACTION="mailto:..."
> in the HTML. Pretty simple thing...
It's a browser thing. I don't believe putting a mailto in a <FORM> tag
is correct HTML...therefore some browsers don't know what to do with it.
...and then there are times when mail is set up wrong...
> I know the technique works because I've received 1500+ form responses
> over the past few months from the Fibblesnork LEGO Survey, the Cool LEGO
> Site of the Week voting, and from this RTL Web Broswer Survey.
Maybe you should put a little notice that if it doesn't work, they
should copy the text from the page and submit the information manually.
It's a pain in the ass for the user, but if they really want to get that
information in to you, they can.
--
-Patrick Delahanty | patr...@internet.com
Obligatory URL: http://sebago.internet.com/lego/
PatrickD
Todd Lehman wrote:
> Probably what I will do is write something that lets anyone put together
> any kind of survey they want, and then host that on my server. More
> general that way, and more fun to write. I'd be curious what kinds of
> LEGO surveys people might come up with...
Simple mailer CGI scripts are available all over the web. You should be
able to find one without trouble. All you have to do is point your HTML
form at it. Most of them have you put your e-mail address in a hidden
field. Others will take it as part of the "ACTION". A walk through
Yahoo should get you some. (I'm suprised you have a web counter (VERY
EVIL!) but aren't using a mailer script.)
...but if you meant a CGI script that lets people put together HTML
forms, that's a royal pain in the butt. I did one for a client several
months ago and it wasn't fun at all.
Joshua Delahunty
In article <329376...@internet.com>,
PatrickD <patr...@internet.com> wrote:
>Todd Lehman wrote:
>> Lou Zucaro <l...@pause.com> writes:
>> > I think something might be set wrong in your script.
>>
>> I wish I knew what. All I'm doing is METHOD="post" ACTION="mailto:..."
>> in the HTML. Pretty simple thing...
>
>It's a browser thing. I don't believe putting a mailto in a <FORM> tag
>is correct HTML...therefore some browsers don't know what to do with it.
>...and then there are times when mail is set up wrong...
Patrick's right. From Graham's "HTML Sourcebook, Second Edition":
"...the ACTION can specify other URLs. For example, in the case of a mailto
URL, the FORM content would be mailed to the indicated address. Most browsers,
however, do not support mailto URLs in the context of a FORM."
Looks like it's time to convert over, Todd. :-)
-- joshua
--
-- Joshua Delahunty (dulc...@alumni.cse.ucsc.edu) RTL CMR/IRL's: 6 --
--
+-------------------------------------------------------------------------+
| Kun det bedste er godt nok |
+-------------------------------------------------------------------------+
Todd Lehman
Joshua Delahunty <dulc...@alumni.cse.ucsc.edu> writes:
>> It's a browser thing. I don't believe putting a mailto in a <FORM> tag
>> is correct HTML...therefore some browsers don't know what to do with it.
>> ...and then there are times when mail is set up wrong...
>
> Patrick's right. From Graham's "HTML Sourcebook, Second Edition":
>
> "...the ACTION can specify other URLs. For example, in the case of a mailto
> URL, the FORM content would be mailed to the indicated address. Most browsers,
> however, do not support mailto URLs in the context of a FORM."
Hey cool -- Joshua's back!
> Looks like it's time to convert over, Todd. :-)
Yup, yup, yup, I know... I don't like it being an e-mail response any
more than anyone else does.
Just a matter of scheduling...
Probably what I will do is write something that lets anyone put together
any kind of survey they want, and then host that on my server. More
general that way, and more fun to write. I'd be curious what kinds of
LEGO surveys people might come up with...
--Todd
Todd Lehman
PatrickD <patr...@internet.com> writes:
> Todd Lehman wrote:
>> Probably what I will do is write something that lets anyone put together
>> any kind of survey they want, and then host that on my server. More
>> general that way, and more fun to write. I'd be curious what kinds of
>> LEGO surveys people might come up with...
>
> able to find one without trouble. All you have to do is point your HTML
> form at it. Most of them have you put your e-mail address in a hidden
> field. Others will take it as part of the "ACTION". A walk through
> Yahoo should get you some.
Aha -- so as a really simple intermediate step, I could use a publicly
available simple mailer script, which would avoid ACTION="mailto:..." --
Neat-O.
> (I'm suprised you have a web counter (VERY EVIL!) but aren't using a
> mailer script.)
I like web counters. Unfortunately, I'm using counter.digits.com, which
I would agree is "evil" maybe because it misses about half the hits. (I
just updated the counter to reflect the actual value according to the
server logs.)
> ...but if you meant a CGI script that lets people put together HTML
> forms,
Yup, I mean a CGI script that lets people create custom surveys so people
can ask questions like "Have you ever had s*x in a pile of LEGO?" and
"What is your favorite Castle sub-theme?" and "Do you think LEGO Consumer
Affairs is smoking dope when they say the 1997 sets won't be out until
January 1, 1997?"
I might give an HTML form interface for making up the surveys, or I might
ask people to submit a text file defining the survey. Haven't decided yet.
> that's a royal pain in the butt. I did one for a client several
> months ago and it wasn't fun at all.
Nah, I don't think it'll be a royal pain in the butt...
The current survey engine I have just reads a text file and generates
all the HTML for the forms and the bar graph responses automatically.
The text file looks like this (line wraps at 80 columns added for news
posting):
===============================================================================
@ part1.html Part I
===============================================================================
-------------------------------------------------------------------------------
$ general General
-------------------------------------------------------------------------------
| toys-percent Of all the toys you buy, own, or play with, what
percentage are LEGO toys?
80-100 Virtually all (80-100%)
60-80 Most (60-80%)
40-60 About half (40-60%)
20-40 Some (20-40%)
0-20 Very few (0-20%)
| brand Which best describes how do you feel about <I>other
brands</I> of plastic building bricks?
strongloyal I wish all other brands would disappear off the face
of the Earth.
loyal I have never thought about trying other brands.
weakloyal Other brands are sometimes tempting, but I know I just
won't get the same quality.
ptui I have tried other brands and I did not like them.
price I buy certain other brands for the discount price.
colors I buy certain other brands for the unique colors.
disloyal I buy other brands and mix them freely with my LEGO.
A brick is a brick.
infidel I buy only off-brands. I don't buy LEGO.
.
.
.
===============================================================================
@ part2.html Part II
===============================================================================
-------------------------------------------------------------------------------
$ obsession Obsession
-------------------------------------------------------------------------------
& ever-beentold Have you ever been told...
play-too-much ...that you <I>play with</I> LEGO <I>too much</I>?
think-too-much ...that you <I>think about</I> LEGO <I>too much</I>?
need-more-hobbies ...that you <I>need more hobbies</I>?
brick-noise ...that <I>digging through a pile</I> of LEGO makes an
<I>annoying noise</I> at 2 o'clock in the morning?
& ever-error Have you ever <I>discovered an error</I>...
box ...on a LEGO <I>box</I>?
instr ...in a LEGO <I>building instruction</I> booklet?
catalog ...in a LEGO <I>catalog</I>?
magazine ...in a LEGO <I>magazine</I>?
& ever-take Have you ever taken LEGO pieces with you...
work ...to work?
class ...to class?
vacation ...on vacation?
bathtub ...in the bathtub?
outside ...in the backyard?
car ...in the car?
bed ...to bed?
dinner ...to the dinner table?
bathroom ...to the bathroom?
.
.
.
-------------------------------------------------------------------------------
$ collecting Collecting
-------------------------------------------------------------------------------
| collector-type What kind of <I>collector</I> are you?
none Not really a collector
casual Casual collector
serious Serious collector
completist Completist (must acquire all in some defined group)
| keep-instructions How do you organize your <I>instruction booklets</I>?
toss Throw them out
box Keep them in the box
pile All in a pile
binder In protector pages in a three-ring binder
other (other)
.
.
.
Making up a form that lets people fill in questions and all the possible
choices shouldn't be tough. But even if it is, it's still too fun to resist.
And I'd much rather be pulling data from DBM files instead of parsing text
files...
--Todd
PatrickD
Todd Lehman wrote:
> PatrickD <patr...@internet.com> writes:
> I like web counters. Unfortunately, I'm using counter.digits.com, which
> I would agree is "evil" maybe because it misses about half the hits. (I
> just updated the counter to reflect the actual value according to the
> server logs.)
If you can access your server log, why even bother with a counter? It's
not like experienced users believe them (or care). There are lots of
sites out there that artificially inflate their count. ...and then the
counter misses a lot too as you've noticed.
They're certainly not for informative information, I just see them as
personal ego boosting and something to slow down the loading of a web
page to a crawl.
> > ...but if you meant a CGI script that lets people put together HTML
> > forms,
...
> I might give an HTML form interface for making up the surveys, or I might
> ask people to submit a text file defining the survey. Haven't decided yet.
I did the HTML form method...more on that below.
> > that's a royal pain in the butt. I did one for a client several
> > months ago and it wasn't fun at all.
>
> Nah, I don't think it'll be a royal pain in the butt...
Yeah...but you didn't actually try to do it yet. (...and you don't know
CGI programming.)
> The current survey engine I have just reads a text file and generates
> all the HTML for the forms and the bar graph responses automatically.
> The text file looks like this (line wraps at 80 columns added for news
> posting):
...
[WAY too long chunk of text deleted]
...
> Making up a form that lets people fill in questions and all the possible
> choices shouldn't be tough. But even if it is, it's still too fun to resist.
> And I'd much rather be pulling data from DBM files instead of parsing text
> files...
Making a form to let people fill in stuff is easy. Making an HTML form
to make an HTML form is hard.
The one the client wanted had no set method. Basically, I had to do a
"What do you want to appear next on your form?" thing and let them pick
"Text", "Radio", "Checkbox", etc... Then they had to choose the number
of answers, then they had to fill in the answers, then they went back
and started over. Creating a form was also a pain in the ass on the
user end.
...but if you're just going to allow users to submit some sort of
formatted text file and then convert that to HTML, that's easier.
--
-Patrick Delahanty | patr...@internet.com
Obligatory URL: http://sebago.internet.com/~patrickd/
Todd Lehman
PatrickD <patr...@internet.com> writes:
> If you can access your server log, why even bother with a counter? [...]
Because I like counters. I think they're nifty. Yeah, they can sometimes
be obnoxious and self-promoting, but I like 'em anyway.
> Making a form to let people fill in stuff is easy. Making an HTML form
> to make an HTML form is hard.
Oh well. Then it will be more challenging.
--Todd
J.D. Forinash
In article <574fco$d...@frazier.backbone.ou.edu>,
Nephilim <je...@teubner.com> wrote:
>But I hate visiting a personal web page and seeing "you are the 27th
>visitor since Jan 15 1996" (or worse, a broken counter).
The worst: "You are visitor <number> since Last Counter Reset!"
If nobody knows when the counter was last reset, why does anyone care?
-F
--
. J.D. Forinash: geek, speed demon, sysadmin, Vogon poet, smeghead
/o\ You're not cleared for that. <IMG SRC = "stardestroyer.asc">
'---` http://www.cc.gatech.edu/people/home/foxtrot fox...@cc.gatech.edu
The more you learn, the better your luck gets. Phoenix:GFW O- <*>