Well, it doesn't make sense except if you have a mail server
that exposes only POP3...
Yes, making the INBOX accessible is exactly what I am looking
for.
Helping James will indeed solve my problem. The remaining
issue is how many mail servers out there expose only POP3 and
don't expose IMAP, and what acceptance do they have /
userbase. According to these number, you can decide if
supporting POP3 (INBOX access) is worth implementing.
My take is that since you have a full IMAP support, you can
very probably support POP3 with minimal changes. I may be
wrong.
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
I do agree with you, there is a lot of POP3 out there, but still:
1. IMAP supercedes and outdates the POP3 protocol
2. In case of webmail everything will be on the server anyway, so
why implement a full mail storage and retrieval mechanism when
it comes for free with an IMAP server?
3. If a service provider only offers POP3, and you deploy the
webmail yourself, then you might also deploy an IMAP server as
well and just retrieve the mail from your provider via a script
utility.
Now you might say that IMAP is not secure, but well, I'd say it has
improved a lot and besides that there are possibilities to shield it
completely (Firewall, secure connection).
I think that it depends a lot on what solution one is searching for.
I propose the following:
1. I will take a look at James IMAP2 proposal
2. We will welcome any effort you might want to invest to achieve
POP3 access :) Jwma is a community effort, working just as
jakarta projects. I am not the one that decides, I am just
commenting and saying that I am not very convinced.
(A word of warning though, I have been working on a proposal
version of jwma based on Struts. This might outdate the old code
base at a given point in time)
3. If I get really really bored, I will take a look at INBOX access
based on POP3.
Regards,
Dieter
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
Just one more comment to make sure we are talking about the
same thing...
1. Technical superiority
You say "IMAP supercedes and outdates the POP3 protocol" as
a reason not to implement POP3, and IMO, that has no
relationship with the issue of supporting POP3 or not.
It is like saying: "HTTP/1.1 is superior to HTTP/1.0" so I will
make my web browser HTTP/1.1 compatible, but not HTTP/1.0.
That'll work, but you deny your users access to HTTP/1.0 only
web sites.
IMO, the final decision of implementing POP3 should come out
of the question: "How many web servers deployed out there
implement POP3 only" and not "is POP3 superior to IMAP"
2. Upgrades
You also say "In case of webmail everything will be on the
server anyway, so why implement a full mail storage and
retrieval mechanism when it comes for free with an IMAP
server?"
It is not always easy to have an IMAP server at hand. Upgrade
or switching to another mail server is not always a
fast/acceptable option.
3. Security
The security of the IMAP and POP3 protocols are
approximately the same if you consider the ability to use a TLS
tunnel for your POP3 connection.
4.Conclusion ;-)
No offense or flame intended, I am just giving my 2 cents here.
I understand the reluctance to implement an outdated
protocol, but that is often what makes a product successfull
bacause supporting a various range of data/servers.
Look at how all the web browsers out there still provide a
strong support for old techs (HTML1, HTTP/1.0, gopher...)
Thanks anyways for all the good work. I look forward into
using your product when a) you implement POP3 or b) James
IMAP implementation will be in a reasonnably stable state.
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
I don't think we are talking about the same thing.
ad 1.
I did not say technically superior, that would depend only on the
implementation.
However, POP3 is for picking up your email to a local client, and
IMAP is for managing your mail on a server. That's WAY different.
Exactly this difference is why we are supporting IMAP but not
POP3.
The IMAP server is handling and storing your complete mailbox,
all operations are done on the server. We just tell the server what
to do, without knowing whether this will affect a file on a hard
drive, a database or whatsoever.
Also it does not matter which client you use to access your mail
store, it will always be the same store, as it is the store on the
server not on any client.
Now this is my argument when it comes to POP3. All it allows me
to do is to access the INBOX to pick up email (and eventually
remove it). It does not support managing a mail store. Thus it
would require a complete store implementation, to do the same as
an IMAP server can do. However, the store will be related to the
application, i.e. it will be only there for a single client.
It's not about v0.9 and v1.1. Is more like a completely different
protocol, for achieving something completely different.
So the final decision to implement POP3 support, would mean that
you have to build a mail store (with all it's mechanisms). Now I
ask you, do you think it makes sense to do so for a single
application, or wouldn't it be better to do this for James, so it
could work as an IMAP server?
ad. 2)
Well see above. It is simply a matter of the huge difference
between what POP3 does and what IMAP does.
It's just about what you really want. A "post box" or a mail store
on the server.
(And again, webmail implicates that the mail store will have to be
on the server).
ad 3.)
I actually meant that IMAP was not widely accepted, because the
first versions were known to be pretty insecure. This has changed
a lot, so the argument does not count any longer.
I agree with the TLS, we actually support that in jwma.
ad 4.)
Well, when you read the above and you understand what I am
talking about, then you probably see that your conclusions are not
exactly related to the original problem.
I checked below, I did not say POP3 is outdated and that is why
we don't support it. Probably you read that between the lines?
If I still did not explain myself well enough, then please let me
know, I really would like to have this straight.
Regards,
Dieter
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=36782
Pierre,
I don't think it makes sense to implement full featured POP3
support, when all the functionality already exists in IMAP servers.
Making the INBOX accessible (the only thing POP3 accounts for) is
probably possible, but the question is if this is what you are
looking for.
I guess the best will be to support James development, helping
them with their IMAP implementation?
Regards,
Dieter
Logged In: YES
user_id=530082
Well, it doesn't make sense except if you have a mail server
that exposes only POP3...
Yes, making the INBOX accessible is exactly what I am looking
for.
Helping James will indeed solve my problem. The remaining
issue is how many mail servers out there expose only POP3 and
don't expose IMAP, and what acceptance do they have /
userbase. According to these number, you can decide if
supporting POP3 (INBOX access) is worth implementing.
My take is that since you have a full IMAP support, you can
very probably support POP3 with minimal changes. I may be
wrong.
Logged In: YES
user_id=36782
Pierre,
I do agree with you, there is a lot of POP3 out there, but still:
1. IMAP supercedes and outdates the POP3 protocol
2. In case of webmail everything will be on the server anyway, so
why implement a full mail storage and retrieval mechanism when
it comes for free with an IMAP server?
3. If a service provider only offers POP3, and you deploy the
webmail yourself, then you might also deploy an IMAP server as
well and just retrieve the mail from your provider via a script
utility.
Now you might say that IMAP is not secure, but well, I'd say it has
improved a lot and besides that there are possibilities to shield it
completely (Firewall, secure connection).
I think that it depends a lot on what solution one is searching for.
I propose the following:
1. I will take a look at James IMAP2 proposal
2. We will welcome any effort you might want to invest to achieve
POP3 access :) Jwma is a community effort, working just as
jakarta projects. I am not the one that decides, I am just
commenting and saying that I am not very convinced.
(A word of warning though, I have been working on a proposal
version of jwma based on Struts. This might outdate the old code
base at a given point in time)
3. If I get really really bored, I will take a look at INBOX access
based on POP3.
Regards,
Dieter
Logged In: YES
user_id=530082
Just one more comment to make sure we are talking about the
same thing...
1. Technical superiority
You say "IMAP supercedes and outdates the POP3 protocol" as
a reason not to implement POP3, and IMO, that has no
relationship with the issue of supporting POP3 or not.
It is like saying: "HTTP/1.1 is superior to HTTP/1.0" so I will
make my web browser HTTP/1.1 compatible, but not HTTP/1.0.
That'll work, but you deny your users access to HTTP/1.0 only
web sites.
IMO, the final decision of implementing POP3 should come out
of the question: "How many web servers deployed out there
implement POP3 only" and not "is POP3 superior to IMAP"
2. Upgrades
You also say "In case of webmail everything will be on the
server anyway, so why implement a full mail storage and
retrieval mechanism when it comes for free with an IMAP
server?"
It is not always easy to have an IMAP server at hand. Upgrade
or switching to another mail server is not always a
fast/acceptable option.
3. Security
The security of the IMAP and POP3 protocols are
approximately the same if you consider the ability to use a TLS
tunnel for your POP3 connection.
4.Conclusion ;-)
No offense or flame intended, I am just giving my 2 cents here.
I understand the reluctance to implement an outdated
protocol, but that is often what makes a product successfull
bacause supporting a various range of data/servers.
Look at how all the web browsers out there still provide a
strong support for old techs (HTML1, HTTP/1.0, gopher...)
Thanks anyways for all the good work. I look forward into
using your product when a) you implement POP3 or b) James
IMAP implementation will be in a reasonnably stable state.
Logged In: YES
user_id=36782
Pierre,
I don't think we are talking about the same thing.
ad 1.
I did not say technically superior, that would depend only on the
implementation.
However, POP3 is for picking up your email to a local client, and
IMAP is for managing your mail on a server. That's WAY different.
Exactly this difference is why we are supporting IMAP but not
POP3.
The IMAP server is handling and storing your complete mailbox,
all operations are done on the server. We just tell the server what
to do, without knowing whether this will affect a file on a hard
drive, a database or whatsoever.
Also it does not matter which client you use to access your mail
store, it will always be the same store, as it is the store on the
server not on any client.
Now this is my argument when it comes to POP3. All it allows me
to do is to access the INBOX to pick up email (and eventually
remove it). It does not support managing a mail store. Thus it
would require a complete store implementation, to do the same as
an IMAP server can do. However, the store will be related to the
application, i.e. it will be only there for a single client.
It's not about v0.9 and v1.1. Is more like a completely different
protocol, for achieving something completely different.
So the final decision to implement POP3 support, would mean that
you have to build a mail store (with all it's mechanisms). Now I
ask you, do you think it makes sense to do so for a single
application, or wouldn't it be better to do this for James, so it
could work as an IMAP server?
ad. 2)
Well see above. It is simply a matter of the huge difference
between what POP3 does and what IMAP does.
It's just about what you really want. A "post box" or a mail store
on the server.
(And again, webmail implicates that the mail store will have to be
on the server).
ad 3.)
I actually meant that IMAP was not widely accepted, because the
first versions were known to be pretty insecure. This has changed
a lot, so the argument does not count any longer.
I agree with the TLS, we actually support that in jwma.
ad 4.)
Well, when you read the above and you understand what I am
talking about, then you probably see that your conclusions are not
exactly related to the original problem.
I checked below, I did not say POP3 is outdated and that is why
we don't support it. Probably you read that between the lines?
If I still did not explain myself well enough, then please let me
know, I really would like to have this straight.
Regards,
Dieter