How should you deal with a popular project that you no longer want to maintain?

I am the maintainer of a project which has a large non-technical userbase. I’ve been maintaining it for about 4 years now and adding new features as they’ve been requested.

I’d like to move on to other projects now and stop developing for this application. Because of the non-technical nature of the users, there have been very few code contributions in the past. I don’t believe I will be able to find anyone else to take over the project in my stead.

Bugs, issues, feature requests – these are still coming in. I am still responding to emails for help, as I am not sure if I should ignore them, tell them that I’m not working on the application, or if I should respond to emails in only certain cases.

What is the best way to ‘abandon’ this project, but still let users use the application?

Update (July 2016) – It didn’t go as planned. I made an announcement in the README and soon after, I started receiving contributions of a more substantial nature. Pull requests with bug fixes, features, documentation, issue activity. Since then, the project has felt ‘reinvigorated’ and I’m now happily maintaining it along with newer projects. I have collaborators as well. At a guess, it may have been the kind of contributions which were affecting my view of the project and with the quality of contributions improving, it didn’t feel like a chore any more.

I’m guessing this is not a project at a workplace where you are a paid employee and something you do in your spare time for free?

If you are making no money from this, then clearly there is no incentive for you, and no incentive for anyone else to come in fresh to deal with it. (unless maybe it is for a charity or similar voluntary organisation)

As an alternative, why not look at the possibility of adding paid for features.

This way you may have some incentive to continue. You might find people willing to pay, especially when the alternative is for the system to stop being actively developed.
(of course people may abandon your system, but what do you care, you already aren’t being paid).

Another option could be to use the project to learn new technologies? Is it a website? Upgrade to the latest technology? Convert from Asp.Net to MVC4 for example? build a mobile version, make it service based and create an iOS app front end for it?

Announce your abandoning of the product to your community of users. Maybe you will find a successor for your role as maintainer. Try to organize some time of handover, as you would with a project in your day job.

As esr put it in The Cathedral And The Bazaar:

When you lose interest in a program, your last duty to it is to hand it off to a competent successor.

Another suggestion for you, which is slightly the opposite of what you’re asking but I think should be in your list for consideration. Have you considered not abandoning it? If you have a project that there are people actively using and have increasing requirements for, but are unable to modify it themselves and you are the only expert in the software… then you are in the position to charge them money for it.

If the source is open, you could consider closing it (your choice if you want to stifle competition developing the project further). When the next feature request comes in, say yes for a charge of $xyz.

Just an option to consider.

It’s a tough thing to abandon your fan base, especially when you’re different from them. If there were developers in the user group, it would be an easy problem to solve: just announce your impending exit and suggest that someone else step up, offering to help them get up to speed before you leave. Since there aren’t, the question is really this: Can you (or your users) find someone to replace you in a timeframe that makes sense to you (or your users).

In the past, I’ve maintained several projects for years longer than I cared to because it was good for my reputation. My stature, comparatively small though it is, in my field has helped me find jobs when I’ve needed or wanted them, and that has value to me. Enough value to make it worth putting my time in when I could. Eventually, of course, I walked away, but I made sure the full project code was available to any successors.

Do you want to abandon the project because…

you don’t want to do it any more ?

Then: Graceful shutdown ala Reader.

Or … hire a developer ( continue below)

Or because you are losing money?

Calculate out-of-pocket costs ( and continue below)

Or because you aren’t making money?

Calculate how much you need to make for you to feel differently :

out-of-pocket costs must be covered
cost of a developer to continue development
a profit margin

Be honest with your users: explain to them that it costs a certain amount of time, energy, etc. to maintain the service.

Then ask for donations and/or charge for existing features. Do not try to invent premium features that just delays determining if the service is really valuable enough to support itself. Just go with the features you have.

If users bitch well then they can go elsewhere. If not enough donations and/or sign-ups, then shutdown.

Be brutal – once you pull the plug don’t look back.

You have a couple of options as other have noted. My option is to put out an end of life notice. Indicate that the product will be shutting down on such and such a date.

Additionally indicate that since this product is nearing end of life, only critical bugs that impact the ability of the application to function as designed or intended will be addressed. I.E. if the server is down you will get it up and running again.

If users have data, you may have to add a way for them to export it.

Take a look at what Google recently did with Reader for guidance. They shut it down and it was a very popular service, but it didn’t fit their long term goals so the tough decision to shut it down needed to be made.

Is some kind of half way measure a possible solution? Keep the project going but lessen your workload?

For example, you say you are still responding to emails for help. Can you set up a users forum and insist all support queries are done through that so other power-users can help?

Well going over to pure paid will kill off a lot of users, but there are a lot of alternatives to pure paid. A video game I play gives donators some extra perks like more downloads per hour “a pure skill based game, not to confuse with pay to win garbage lol”. Another game Path of Exile offers cosmetic upgrades. Other sites put surveys in exchange for bandwidth. Donation coder gives free users a licenses for (X Time) renewable many times as they want, but donators get a permanent licenses.

There are tons of options offer it for cash but still keep the free users around as well.

Most people have no problem supporting something they like so honestly I’d just try asking first by setting up a tip area calculated to cover your monthly cost.

You have included the open-source tag, so I guess your project is software is open source.

there have been very few code contributions in the past

That’s unfortunate, but understandable in case you were doing everything. Many users don’t get involved as long as it works reasonably.

Some leaders like to delegate all responsibilities, and some leaders like to keep tighter control. While a balance is necessary, delegating as soon as possible is the key here.

I have created over 30 open source projects, and many are still active even though I have left them. Here is what I would recommend:

Give bug tracker access VERY generously, maybe to anyone who has ever contributed a line of code. If someone starts doing crazy things (very low probability), you still have admin control to remove them. Don’t forget to give out other rights: source code control, wiki, crowd-translation, facebook page, twitter account, official website, google analytics, etc.
Post in the forum (and website notice) announcing that you are retiring, and looking for a new project leader.
Even if no one steps in as a team leader, fatal problems that could occur (silly example: a hard-coded URL becomes 404, making the app crash at startup), it will probably be fixed by someone. If nobody fixes fatal flaws, then it means you should not worry too much anymore, you have done what you could, but the project does not appear to be viable anymore.

Filed under: softwareengineering - @ 13:10

Thẻ: maintenance, open-source, project

Thiết kế website giá rẻ

Danh mục

How should you deal with a popular project that you no longer want to maintain?