[At-Large] Google Developers : Humans can't read URLs. How can we fix it? - HTTP 203
Dev Anand Teelucksingh
devtee at gmail.com
Tue Feb 11 03:08:22 UTC 2020
Dr. Alejandro Pisanty Baruch <apisan at unam.mx> wrote:
"is this just one more iteration of "domain names will cease to be relevant
soon" or is it a move or trend that will definitely undermine their
importance and therefore also change the markets?"
I think no, this is not an iteration about domain names becoming
irrelevant. The discussion focused on how to ensure browser users can
what domain they are visiting so IMO, it strengthens the use of domain
names by having users trust the domain they are visiting.
The challenge as I see it is with the prevalence of many new TLDs, the risk
of phishing and malware attacks are higher where bad persons try to trick
persons to either:
- go to bad persons' website(s) at domains pretending to be another
organisation/service in order to get user login credentials to such
- open links to malware files and become infected.
So typically, bad persons would use typos that look similiar to the
legitimate domain or add a few characters before or after the 2nd level
So for example.com, there could be exannple.com , examp1e.com, eexamples.com
and examples.com .
With more TLDs and wider use of ccTLDs, bad persons could also do things
like examples.xyz, examp1e.tv and so on.
With longer URLs, one can mask the legitimate domain as a third or fourth
level subdomain of a bad persons' domain.
Very long URLs like this are more difficult to parse and read on a mobile
phone. Content Management Systems for websites over the years unfortunately
generate long URLs that do work to direct persons to websites
but are obscure for persons to parse.
So an example of a bad URL :
Persons glancing at this URL would see icann.org and may think everything
is fine and go ahead and click on the URL when its not.
Nearly all browser vendors are using the Public Suffix List at
https://publicsuffix.org/ which started as a Mozilla Initiative.
>From the Public Suffix List website, a "public suffix" is one under which
Internet users can (or historically could) directly register names.
Tracking this is important to ensure that a website can't set cookies that
other websites can read or modify under the same domain.
What the episode showed was that Mozilla Firefox (at least on the desktop)
highlights the part of the domain that is the public suffix by making it
slightly bolder than the rest of the URL.
It is however to me, very, very subtle and I wish that Mozilla (and other
browsers) make the public suffix part more bold or even a slightly bigger
font size to alert users as
to the true domain being viewed.
Some mobile browsers are showing only the domain and not the rest of the
URL and allow for the full URL can be viewed and edited when you tap on the
URL bar which also helps.
Dev Anand Teelucksingh
On Sun, Feb 9, 2020 at 3:23 PM Dr. Alejandro Pisanty Baruch <apisan at unam.mx>
> thanks a lot for sending this. While at first it may seem off-topic, it
> cuts deeply into very relevant questions for us: is this just one more
> iteration of "domain names will cease to be relevant soon" or is it a move
> or trend that will definitely undermine their importance and therefore also
> change the markets? If it proceeds, what balances of power will appear that
> we don't face now? One example: will the companies that operate search or
> that provide browsers (or both!) earn new powers? This is already being
> discussed elsewhere. What are the views that are relevant for ICANN
> Internet users At Large, and other stakeholders?
> Alejandro Pisanty
> - - - - - - - - - - - - - - - - - - - - - - - - - - -
> Dr. Alejandro Pisanty
> Facultad de Química UNAM
> Av. Universidad 3000, 04510 Mexico DF Mexico
> +52-1-5541444475 FROM ABROAD
> +525541444475 DESDE MÉXICO SMS +525541444475
> Blog: http://pisanty.blogspot.com
> LinkedIn: http://www.linkedin.com/in/pisanty
> Unete al grupo UNAM en LinkedIn,
> Twitter: http://twitter.com/apisanty
> ---->> Unete a ISOC Mexico, http://www.isoc.org
> . . . . . . . . . . . . . . . .
> *Desde:* At-Large [at-large-bounces at atlarge-lists.icann.org] en nombre de
> Dev Anand Teelucksingh [devtee at gmail.com]
> *Enviado el:* viernes, 07 de febrero de 2020 11:21
> *Hasta:* At-Large Worldwide
> *Asunto:* [At-Large] Google Developers : Humans can't read URLs. How can
> we fix it? - HTTP 203
> "In this episode, Jake makes the case that URLs are impossible for humans
> to interpret, especially when it comes to security. What are browsers doing
> today to overcome that? And, is there a better way"
> Watch the 20 min episode :
> Dev Anand
-------------- next part --------------
An HTML attachment was scrubbed...
More information about the At-Large