Thanks Janos
I'll test further on your piler EE 1.1.0 when it's out and get back
to you with more feedback. Thank you so far for what you have done!
I'm looking forward to seeing Piler in the Enterprise market.
Regards
----- Original Message -----
From: "Janos SUTO" <[email protected]>
To: [email protected]
Cc: "Sam Gelbart" <[email protected]>, "David Jacobson ([email protected])" <[email protected]>
Sent: Thursday, 20 February, 2014 3:12:34 PM
Subject: Re: some future plans for piler, and a 10 question long survey
Hello Darryl,
I'm working on the binary version of piler EE 1.1.0, and finalizing
some settings with the postinstall script. When it's done, then I'm
going to provide a 64 bit package for both debian 7.x and centos 6.5,
these will be the supported platforms for piler EE.
Give me a few more days until this release is completed. Then it will
be available along with a trial licence for some time.
The GUI based provisioning already works. You create an ldap entry,
then the domain(s) for the given customer, and finally map a customer
id [a-z0-9_] to the primary domain. Note that it's run from crontab
in every 5 mins, so expect a max. 5 mins delay in the provisioning
process. I'll also add a status info the customer page, eg. "pending"
and "created".
The per domain separation is ready. I slightly tweaked the attachment
storage, so attachments are stored under /var/piler/astore (and not
in /var/piler/store).
I've also compiled a TODO list of the survey feedback, however some
of the requested features are for 1.2, and 1.3 versions of piler, so
I'd not add any new features before completing the very first version
of piler EE (which should be considered as a beta) giving you an
opportunity to provide feedback, and for me to see where it needs
some additional polishing, etc.
If you have something different in mind, then let me know, I'm open
to suggestions.
Janos
On 2014-02-20 12:41, Darryl Sutherland wrote:
> Hi Janos
>
> For now, what is the way forward? Shall we await the SaaS release you
> mentioned a few
> posts ago and take the discussion further from there?
>
> Regards
>
> ----- Original Message -----
> From: "Janos SUTO" <[email protected]>
> To: [email protected]
> Cc: "Sam Gelbart" <[email protected]>, "David Jacobson
> ([email protected])" <[email protected]>
> Sent: Monday, 17 February, 2014 10:41:38 PM
> Subject: Re: some future plans for piler, and a 10 question long
> survey
>
> Hello,
>
> thanks for both of you to share your thoughts. You both raised very
> valid points.
>
> On 2014-02-17 14:27, Darryl Sutherland wrote:
>
>> It's good to hear some community feedback on this, so thanks for the
> > post.
> >
> >> One could follow an example from mailborder or mailarchiva or
> magento
> >> they
> >> placed their ee edition under a beta with a trial license (free)
> once
> >> the
> >> product reached stability and a complete feature set +/- 12months
> the
> >> product became commercial only.
>
> this sounds like a viable approach.
>
> > I have experience with Mailarchiva, and it's trial versions are not
> > "free". You
> > have a limited time in which to use the product, where after it's
> > features are
> > locked and you have to pay to unlock the features you want.
> > Thereafter you find
> > yourself hopping from one broken version to the next, trying to fix
> > bugs by
> > upgrading to the "latest stable release" complete with it's own
> > broken code.
> >
> > This is the kind of scenario I would like to see Piler avoid -
> where
> > features are
> > more important that application stability.
>
> I think it's possible by supporting a given version of piler EE for a
> year or so.
> Currently I don't backport any fixes to the previous stable CE
> releases, only adding
> them to the current master branch. However with an EE release, eg.
> 1.1.0 there could
> be 1.1.1, 1.1.2, ... as bugfix releases. Once a version (1.1 in this
> case) is released
> then no new features are added, only bugfixes, so there won't be any
> new bugs (well,
> at least not in theory).
>
> > We are not looking for a free ride - we just want a reliable,
> stable,
> > scalable
> > enterprise quality solution and are happy to pay to use it - as
> long
> > as it is
> > providing a stable, well tested code base.
>
> that's fair enough
>
> > imply. That being said, we are open to discussions around funding
> of
> > Piler
> > development.
>
> ok, we'll discuss the details, and I'm open to suggestions.
>
> >> The only real people that get hurt buy the ee edition are smaller
> >> companies,
> >> however the ce edition would still offer enough value at the price
> >> point even
> >> if it lacks features.
>
> well, I intend to keep improving the ce edition, and it may have all
> the necessary features for archiving emails. However I believe that
> some features may be more useful to bigger companies with more users,
> so perhaps the smaller companies are not hurt (that much) by not
> offering them some features they wouldn't use most of them.
>
> Eg. an SMB company may not be interested in the accounting feature,
> while
> a service provider may appreciate it. So it seems to make sense to
> put
> the accounting feature to the ee edition only. Note that this is just
> an example, I won't remove any current feature from the ce edition.
>
> Or another example could be the multitenancy feature. Probably an SMB
> company doesn't need it at all. Anyway, I won't stop anyone to use
> piler
> ce, and to accomplish an ee-feature with it in creative ways.
>
> > I wouldn't like to see anyone getting hurt by the EE edition, which
> is
> > why
> > I would like to see Piler retaining a free to use (CE) edition.
> Right
> > now,
>
> Don't worry, I'm going to keep the ce version, since I'm a great fan
> of
> open
> source products.
>
> > This cash injection would allow the mailpiler project to hire more
> > developers and grow.
>
> indeed, there are some skills I could use to improve piler.
>
>>> On 17 Feb 2014, at 1:24 PM, Darryl Sutherland <[email protected]>
> >> wrote:
> > >
> > > As such, I would recommend that we consider the existing changes
> as
> > initial
> > > development of the CE edition. Once CE multitenant has reached a
> > reasonable
> > > stage of stability and development, we can take the stable code
> and
> > package
> > > it as the EE offering which will generate revenue for you.
>
> ok, we'll see if it would work out.
>
> Best regards,
> Janos
