Oooh! I've been waiting for that. Still haven't found anything sufficiently clearly better than Ameol to tempt me away from it so far. Mozilla's mail&news support works close enough to how I do to look very tempting, even though kjersti (An Agent/Eudora user) shudders with revulsion. I shudder with revulsion at Agent, and while I can and do use Eudora, it's a bizzare mixture of Byzantine complexity and plain ol' clunkiness.
Sod Imap, though. Server storage requirements are astronomical these days; it's sometimes awkward and nonstandard to configure; it's not quick; I do not want to trust my ISP to hold my mail store, thankyou; and POP3 is so widely supported it will be a VAST task to move across. Nice idea, ain't gonna happen soon.
You want the ISPs to store years worth of mail, not days worth. That's terabytes, petabytes of storage. It needs new infrastructure, new backup requirements, all sorts; it's a hell of a lot more than a few NAS units.
And what is access to big file attachments going to be like over a 56K modem, eh?
Not really - we specced out the webmail system for Freeserve a few years back, and even then 3TB of redundant storage came in less than £30K.
Then you can save even further by going hierarchical.
The problem with ISPs is that they rarely treat their businesses as enterprise IT. If they did, they'd improve margins no end. And I speak as someone who designed/used to run tech for a national ISP...
Fair points, tho' it doesn't address the bandwidth/responsiveness question. But [a] they've got to offer POP3 as well and [b] persuade users to change to IMAP.
Heh, yeah, they move pretty fast. Supersonic goalposts...
But I'm unsure why you raise the attachments issue - one of the advantages of IMAP is the ability to see that there are attachments present before downloading... so allowing you to download attachment-laden messages when convenient.
Unlike sitting there wondering why a POP3 download was taking ages.
This is true, but it switches the onus of understanding back to the user again, which is a bad thing. We're trying to make these machines easier and simpler!
Getting Joe Q User to understand local vs. remote storage and what an attachment is, distinct from the message itself, is - let's just say "not a desirable task".
I think you're slightly under-estimating end users here - and the evolution of ISP technologies.
For one thing, web mail services like Hotmail are already showing people that their mail can be left on servers, and that attachments can be downloaded at leisure.
Hmmm... hey, Simon, I work for an ISP (well, it's also an ISP) in the US. D'you think you could clarify that comment? I know I have some bosses who'd like the idea of the company saving money.
Ah, that's the $1000/day question! (or at least it is when I have my consulting hat on).
Key issues (and this is off the top of my head, I hasten to add)I think are:
1) removing legacy and long-term "quick-win" solutions and replacing them with redundant n-tier systems. NAS and transparent caching should be used where possible.
2) investing in core internal networks - upgrading authentication and migrating to central LDAP from RADIUS. Implementing management solutions such as Netcool, to enable proactive maintenance. Redundant systems will reduce downtime here, too.
3) stick to open standards where possible.
4) join GRIC or similar
5) begin to add usage monitoring to all services, in order to find uneconomical users and move them to alternate pricing
6) add support for mobile devices. This is the opportunity to prevent mobile operators from acquiring customers. It should be possible for a user to buy a 2.5G or better device and be able to use your services with only minimal configuration.
7) add billing systems capable of managing multiple services. Time-based billing and flatrate are commodity business models (look at AOL for an example of where these fail!). ISPs need to move away from the commodity market and find value-adds that can be used to make money... Lessons can be learnt from the behaviour of mobile operators in saturated markets, such as in Hong Kong.
![[personal profile]](https://www.dreamwidth.org/img/silk/identity/user.png){kind=small}
(no subject)
Sod Imap, though. Server storage requirements are astronomical these days; it's sometimes awkward and nonstandard to configure; it's not quick; I do not want to trust my ISP to hold my mail store, thankyou; and POP3 is so widely supported it will be a VAST task to move across. Nice idea, ain't gonna happen soon.
Re:
Which is ridiculous when you consider I paid £900 for a 4GB hard drive only 8 years ago.
And if you're deploying NAS, look at the cost of a fully configured Xserve RAID or an Iomega NAS server.
You're missing about 5 years of price reductions in your rebuttal :-)
(no subject)
You want the ISPs to store years worth of mail, not days worth. That's terabytes, petabytes of storage. It needs new infrastructure, new backup requirements, all sorts; it's a hell of a lot more than a few NAS units.
And what is access to big file attachments going to be like over a 56K modem, eh?
Re:
Then you can save even further by going hierarchical.
The problem with ISPs is that they rarely treat their businesses as enterprise IT. If they did, they'd improve margins no end. And I speak as someone who designed/used to run tech for a national ISP...
(no subject)
Re:
(no subject)
And now we're moving from "IMAP-compatible client" to "good IMAP client with caching." Just watch those goalposts go! :¬)
Re:
But I'm unsure why you raise the attachments issue - one of the advantages of IMAP is the ability to see that there are attachments present before downloading... so allowing you to download attachment-laden messages when convenient.
Unlike sitting there wondering why a POP3 download was taking ages.
(no subject)
Getting Joe Q User to understand local vs. remote storage and what an attachment is, distinct from the message itself, is - let's just say "not a desirable task".
Re:
For one thing, web mail services like Hotmail are already showing people that their mail can be left on servers, and that attachments can be downloaded at leisure.
(no subject)
(no subject)
Re:
Key issues (and this is off the top of my head, I hasten to add)I think are:
1) removing legacy and long-term "quick-win" solutions and replacing them with redundant n-tier systems. NAS and transparent caching should be used where possible.
2) investing in core internal networks - upgrading authentication and migrating to central LDAP from RADIUS. Implementing management solutions such as Netcool, to enable proactive maintenance. Redundant systems will reduce downtime here, too.
3) stick to open standards where possible.
4) join GRIC or similar
5) begin to add usage monitoring to all services, in order to find uneconomical users and move them to alternate pricing
6) add support for mobile devices. This is the opportunity to prevent mobile operators from acquiring customers. It should be possible for a user to buy a 2.5G or better device and be able to use your services with only minimal configuration.
7) add billing systems capable of managing multiple services. Time-based billing and flatrate are commodity business models (look at AOL for an example of where these fail!). ISPs need to move away from the commodity market and find value-adds that can be used to make money... Lessons can be learnt from the behaviour of mobile operators in saturated markets, such as in Hong Kong.
Re: Groan!
Then you point out that the next thing is of course a two-week gap analysis, followed by project definition.
And then you introduce the rest of the team...