Using GET in my opinion would make it easy for users of the
site to create hyperlinks to whois information on domains.
Something similar to what whois.sc does.
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
I know there are advantages and disadvantages for both GET
and POST. Personally I like POST because the short URL
(although you can't bookmark or link to specific domain
lookups). But we can add an option to select GET or POST
at the web based installer.
Also, I think it can be easier for spam harvesters to use
GET (I'm not sure). And search engines might also use GET
links and create more load on the system. So in the
meantime let's leave it with POST, until we have a web
based installer and then each webmaster will select for
himself.
By the way, why would users want to bookmark or link to
specific domain names? I thought users need to check any
domain name only once, not regularly.
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
It's a good idea to have the administrator choose whether to
use POST or GET. Spam harvesters can use POST or GET. I
suggest we also include some obfuscated registration code
like what is available from the CAPTCHA Project
(http://www.captcha.net/) that has to be entered into a form
field before carrying out the lookup. This would go a long
way to preventing automated queries. If we do decide to
implement this, only using POST would make more sense.
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
The disadvantages of turing tests, such as CAPTCHA, are:
- It takes more time and annoys the users
- Some people can't pass the tests (for example, disabled
people)
For this reason, I don't want to include a turing test in
Speedy Whois. I know there is the risk of automated
queries, but I don't know how bad is the risk. I think
most quality whois websites don't use turing tests, for
example:
www.whois.sc
www.whois.ws
www.iwhois.com
Maybe they use other methods against automated queries.
If they do, we might consider using similar methods.
Uri.
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
Logged In: YES
user_id=539487
Using GET in my opinion would make it easy for users of the
site to create hyperlinks to whois information on domains.
Something similar to what whois.sc does.
Logged In: YES
user_id=1264877
I know there are advantages and disadvantages for both GET
and POST. Personally I like POST because the short URL
(although you can't bookmark or link to specific domain
lookups). But we can add an option to select GET or POST
at the web based installer.
Also, I think it can be easier for spam harvesters to use
GET (I'm not sure). And search engines might also use GET
links and create more load on the system. So in the
meantime let's leave it with POST, until we have a web
based installer and then each webmaster will select for
himself.
By the way, why would users want to bookmark or link to
specific domain names? I thought users need to check any
domain name only once, not regularly.
Logged In: YES
user_id=539487
It's a good idea to have the administrator choose whether to
use POST or GET. Spam harvesters can use POST or GET. I
suggest we also include some obfuscated registration code
like what is available from the CAPTCHA Project
(http://www.captcha.net/) that has to be entered into a form
field before carrying out the lookup. This would go a long
way to preventing automated queries. If we do decide to
implement this, only using POST would make more sense.
Logged In: YES
user_id=1264877
The disadvantages of turing tests, such as CAPTCHA, are:
- It takes more time and annoys the users
- Some people can't pass the tests (for example, disabled
people)
For this reason, I don't want to include a turing test in
Speedy Whois. I know there is the risk of automated
queries, but I don't know how bad is the risk. I think
most quality whois websites don't use turing tests, for
example:
www.whois.sc
www.whois.ws
www.iwhois.com
Maybe they use other methods against automated queries.
If they do, we might consider using similar methods.
Uri.