The future of SME Server

DRAFT, WIP

A document to discuss potential management and roadmap at SME

First, in case you wondered, a bit about me. My name is John Crisp, I am 50 (dammit, I hate that number), British with a very sick sense of humour, but live in Spain where my wife (The boss both literally and metaphorically) and I run a small business back in the UK.

I'm a jack of all trades and master of none. I am not a computer professional, though I have been using computers one way or another for over 30 years. I run the IT for our little company, and the reason for my interest and involvement with SME is down to the use of SME servers in the business and friends for around 10 years. I am not a developer. Just a tinkerer really.

SME has been the bedrock of our business and I have, over the years, tried to put something back in to the community. Now seems the time to put a bit more back.

I never intended to be the instigator of current events. Things seem to have fallen on my shoulders by chance rather than by design. People said 'just do something'. So I decided I would try. I don't have all the answers. The future is not my decision, but those of the community. I see myself currently as a conduit and driver of these ideas. I am not here to make any decisions, but to try and get a consensus and put it in motion, to harness the potential that is here, and try to make it work. I have no axes to grind with anyone. I speak as I find, and may occasionally be blunt, but intend no one any malice.

As many are well aware, SME is in potential turmoil at present. The decision by Filippo Carletti to effectively fork SME has had huge implications for this community.

Since the attempt at forming some sort of management failed in 2009, the distro has effectively been leaderless. Getting SME V8 out of the door was a massive task, and an even bigger achievement. The work of numerous people made this happen, and everyone of them should be congratulated on such a fantastic job.

V8 was a target, and with it the community, as it was, had a focus. That drove the development. But at that point, the roadmap ended. Filippo decided that he could see no future in the project and decided to take the matter in to his own hands.

With any project or business there has to be two core elements. A roadmap or plan, and leadership and management to see that the plan is carried out. Boring stuff, but completely necessary. Filippos decision was born out of necessity. He made some plans and has the organisation to carry them out.

Management, and plans. The two are inextricably linked. One cannot function without the other. A symbiotic relationship.

At the moment SME has neither. If we want to move SME forward we have therefore got to build both of these at the same time.

First, a plan

I have been trying to canvas opinion from users on the mailing lists and via IRC. I have also received personal emails as well.

I have had mainly good comments, but some criticsing the project for having no leadership or direction. I have been a bit short with those who cricised, and want a lot, but have done nothing themselves to help. “I'm too busy selling SME servers to do anything to help” has gotten short shrift.

Although not a massive sample, the general consensus seems to be as follows :

Move the current base to CentOS 6 as soon as possible

This is to allow installation on newer hardware, allow easier installation of other applications with less intereference of the existing subsystem. CentOS 5 is already getting on. Users want to keep abreast of newwer technologies. The pace of change is fast. We may all like it, but we cannot change it. We have to make some for of attempt to keep there or thereabouts. Users and developers want the knowledge that the platform will be current and supported to allow them time to develop applications for it.

Current status : People are already working on porting SME to CentOS 6. It has been referred to as SME Server v9. SME v8 to be maintained for the foreseable future, but the focus of development and effort will be on the new system to try and get it out as soon as possible.

Need to increase funding

Site costs - all donations go there today, and it is not enough. Shad has been open and provided what numbers he can. Development money for Charlie, Development money for Shad If we generate funds then there will be a need to account for them and means accountants and lawyers

Traditionally SME has been developer led, but it also needs a firm financial footing to help fund current and future costs. No money, no honey as they say.

Current status : Shad has already reported that average income is only a couple of hundred dollars a month and that just about meets his hardare costs.

There are suggestions on how to monetise SME whilst still retaining its Open Source tradition. The first thing that springs to mind is some form of advertising on the web site. The current site, though a great resource for SME, does not really sell the product. In the first instance there is a lot of material that is very old and needs bringing up to date – there is work on going already in that department. There are a number of people looking at ideas on how to get greater visibility, more users, and attract potential sponsors or advertisers. We need to have a product in which people see a future, and a concept they can buy in to.

There are a small group looking at the whole marketing aspect of SME including redevelopment of the front page. There have been a varity of suggestion that are being looked at :

Should we rebrand or just redefine SME in the market place ? • revitalize/redesign the website • make it sexy • look at (appropriate) advertising • Donations • Subscriptions • Would we use a CMS or do a ground up design ? • If CMS the which one (So many to choose from) • Run a contest on Contribs asking for ideas for an updated SME brand ?

Other suggestions include asking for specific donations of hardware, or assistance with media, marketing, website design, finance and law.

It is excellent to have the the knowledge that we have “A Linux server OS that's had 11 years to improve” but to sell it, the marketing needs to reverse that and say something like “We have a great new product in development backed by 11 years experience.” We need some positive 'spin' on things, some publicity, and need to attract some attention.

Love it or loath it, Social Media is in. We don't all have to embrace it, but we DO need to use it to get the message out there.

SMEs greatest feature, and its greatest downfall, is its simplicity. It tends to attract less technical users who do not feel able to help with coding or more technical matters. We have to convert the 'I can't help' mentality by asking different questions. There are dozens of different people out there and they can all bring something to the party. Having chatted with quite a few, quite often you find that they either just need some encouragement and a a bit of confidence, or that they other skills that are useful.

This needs developing and working on.

Another thought was schools/ universities – trying to get younger people involved in the project. After all, they are the future.

I'm sure there are plenty of other avenues that can be looked at.

Need for some form of organisation

Whether we like it or not, we need to have some form of organisation. Without it the project will be like a headless chicken.

Currently SME is not big enough to support a massive management structure.

In discussions it seems that a simple structure similar to the following would be a reasonable path to take :

• Development
• Bugs & maintenance
• Contributed packages (current and old/ unmaintained)
• Wiki & Docs
• Forums
• Media / Marketing
• Finance & Legal

We have a number of people who have indicated they will assist. The list is by no means exhaustive and I have missed people off – I still can't always remember peoples names couple with email addresses and IRC nicks ! If I have missed you, please add yourself.

• Current top of the pile : John Crisp

• Development :

Ian Wells is currently leading the charge in this area. He would like around 7 developers plus a few testers to help verify.

Ian Wells Charlie Brady Shad Lords Greg Zartman Daniel Berteaud John Crisp

• Testing :

jester ?? Huib de Visser

• Bugs and maintenance

Contributed packages : Package maintainers

Old packages : Jean-Philippe PIALASSE

• Wiki & Docs

Chris Burnat Terr Fage stephane de Labrusse (stephdl)

• Forums

I do not yet know who the moderators are but I would be grateful if someone could put their names here !

• Media, Marketing, Legal & Finance

John Crisp Greg Zartman Tony Keane