Open alpha vs Closed alpha?

[eugeneloza]

I was wondering, which way is better? Open alpha or closed alpha?

Closed alpha:
I.e. there is a starting opensource game project. There even might be source on github or somewhere else, but no dev snapshots / releases. I.e. without proper compilation skill its closed to the public which only sometimes gets screenshots or videos.
Until the game looks well and is playable.

Versus

From the very beginning there are dozens of dev.snapshots each can download and run.

On one hand, open alpha gives some access to early feedback & interest in the game, stimulating the development. On the other hand as far as I see, practically there is almost none - the feedback mainly comes from people I've asked to play and tell me what they think. I get almost the same amount of feedback if the alpha would be closed.
Open alpha may disappoint users due to low quality graphics & glitchy gameplay and make them loose interest in the project before it reached anything. Most of the project downloads arise at the week the project info was published at some site.
And if someone is really interested in the project I can simply upload a snapshot for him in return for promise to give the feedback.

The main problem is lack of feedback. First, feedback is impetus of the free project development. You see it's interesting, you see someone likes it or not. Second, you see what you've might have missed in the gameplay or tutorial/help. And third its bug report, saves you from having to play hours to hunt bugs yourself and aids detecting different platform compatibility.

So, which one is better? Closed alpha and invited testers, or open alpha and hope for the future?

[Vandar]

I think it's good to have some friends to whom one can send early builds and reliably get feedback from.

But I use to offer builds of my projects as soon as there is something to show. I hope to get more feedback this way, but there are so many people competing for attention these days, it hardly happens - but it doesn't seem to do any harm either.

[andrewj]

I think it depends on how long it will take to make the fully playable version of the game.

If you can develop it in under a year (two max) then being open the whole way can work, people can stay interested and see rapid progress in the game.

If it is gonna take a really long time, like 5+ years, then I think you want to stay under the radar and only "surface" when the game is nearly ready, and you can wow everybody with it. And if it fails, as most projects do, then nobody is getting upset except yourself.

Feedback can be good for motivation, but I generally find that people rarely say anything which I did not already know or had not already thought of. Yes I know that menu is ugly! Yes I know it crashes if you turn left too fast! Yes it would be great if the monsters had better pathfinding! (Maybe this just means I have been developing software too long )

[eugeneloza]

take a really long time, like 5+ years

Well... yes... really long time

people rarely say anything which I did not already know or had not already thought of

Sometimes yes... but for me it's a little different experience. As one of the commenters of Project Helena told that he couldn't beat a long corridor... It was just then I realized that there was absolutely no hint on enemy behaviour (i.e. the long corridor is most often the easiest thing to beat)... I've got some questions 'how to do...' or 'I wish I could...' which were implemented, but I've forgot to mention them in help...
And just... yes... When I read a comment I understand I'm not just doing that for myself to have fun. Other people might have fun too
Often I'm sooooooo lazy... And feedback sometimes (but, unfortunately not always) stimulates to work.
E.g. I've completely lost recent 4 days just because I was sitting and doing nothing in particular (playing Endless dungeons or reading random articles around internet). I feel really bad about it... I could have done over 2000 lines of code. Or a few 3d models. Or a trailer. Or voiceacting...

[Vandar]

It depends on the person. I often noticed that my ideas are a good start, but by discussing them with other people they can be improved quite a bit. That's why I consider it important to talk about game design, and for this, a group of friendly and trusty people looks important to me (and I'm lacking discussion partners these days ...)

[Jastiv]

I'm not really sure that closed alpha has any advantages. For a long time (as in over 10 years.) my project had no releases, not because it was unplayable (it was, it was actually a hard fork of another project) but because I wasn't happy with the missing/ugly graphics, crummy game mechanics and bugs.
Then I decided I was being stupid and I should really, ready or not make a release. I made a release of the pre-alpha version right before I removed it from the cvs. I rearranged some files, changed some graphics and then made an official alpha release moving the latest to git version control (no, I did not put the ten+ years of history in git. although you can still take the old stuff out of cvs, Sourceforge discontinued cvs, so no more commits will be made to it., so now it has two release, a pre-alpha and an alpha. I don't know when I will make another release, probably sometime before this year is out, or early next year, since there have been so many changes from the 0.1.0
I don't know if that was the best choice, but I decided I better do something.