Have you ever submitted code that obviously doesn't work? And why do we need alpha versions when you can think well and do well right away?

Question

Work organization

At work, we need to roll out an alpha version of a new module for a small project.
Last time we did MVP and it was still somehow clear: we abandoned part of the functionality so as not to do it badly and with possible errors
. And this time we are doing the alpha version. Here we are doing the functionality that we don’t want to do, and we’re just doing it incompletely, for example, a form without validating some fields (and server-side too, yeah).

And then, probably, it will be necessary to release another version, eliminating and redrawing some of the problems.
And then more.
Some kind of approach "just not to sit idly by."
Isn't it better to just sit back and slowly make a normal product right away?It seems that it will not be so much longer even compared to the rollout of the alpha (!!!), because a good thought out for several man-days will allow you to think in advance and discuss all the points that would have to be "investigated by typing" in the process of creating version by version. And the fact that no one knows how, it's just human stupidity. It is necessary to develop more such a quality as perfectionism - it fundamentally will not allow something to be redone, because traces remain from the alterations, and this is no longer an ideal. Wouldn't that save more resources than it wasted?

Answer 1

What does "doesn't work" mean?
Do you have a list of features that should be in the alpha version, but you didn't make it, or is it buggy/falling? Then it's just a bad job.
If everything that should be for the alpha version is done and works, then everything works.
But how to plan releases, product development, when and what to roll out - these are decided by the products. If you have questions and you think that they somehow did everything badly, then first of all these questions should be asked to them. Find out why and what is the purpose/task of the alpha version.
You asked why this alpha? what goals does this release pursue, what tasks do they want to close?
Alpha / beta and other versions of product testing were invented not to "x-to x-to and in production",
Maybe you really have stupid products that just strain developers to "at least do something", then plug holes, do not want to think, and generally do outrageous things and only you can save the product with the help of an army of toasters, or maybe vice versa It's something you don't understand or don't know.
Have you checked both options accurately and reliably?

Wouldn't that save more resources than it wasted?

No. This will consume 100 times more resources. Just because you do not have the necessary information to think through everything and make it "perfect". So the part you just make up, consciously or not. and the more fully and in advance you try to "think through all the moments in advance," the more fantasies you will have. And then, when the system is launched in the real world, it turns out that in fact everything is not as you thought, plus half of what you really guessed during this time has already changed and thousands of man-hours will be flushed down the toilet.
Perfectionism is, firstly, a serious personal cognitive problem that needs to be addressed, even regardless of working moments. This is the illusion of correctness, which paradoxically leads to the fact that in reality you make everything worse.
And secondly, he ruined so many products / projects that you can’t imagine.
Get rid of perfectionism, you can do really good and high-quality things, no matter how strange it may sound.

Answer 2

It's just a technical name: alpha - testing by the developer company, beta - testing on the side (therefore, it is often not sold, but given out for free under the terms of error messages)

Answer 3

Read about Lean Startup https://ru.wikipedia.org/wiki/%D0%91%D0%B5%D1%80%D...
Between two extremes - carefully plan and release a new product every ten years (when it already either morally obsolete, or all competitors have already divided the market) and releasing a new non-working product every week is a golden mean. It remains only to understand where it is :-)

Answer 4

There are certain requirements for alpha, beta and full release.
If the alpha requirements are met and the alpha is scheduled, then it should be output.

And then, probably, it will be necessary to release another version, eliminating and redrawing some of the problems. And then more. Some kind of approach "just not to sit idly by."

In addition to your specific piece of work that you want to do perfectly, there are other people who may need to look at the alpha version, and start doing something already based on it.
Check the integration, see if something needs to be changed, not wondering how it will be, but after seeing some version in action, write autotests for the forms, take screenshots and show it to the customer. It is possible that such a clause is in the contract, and money is paid not just for the final product, but for compliance with all the requirements specified in the contract, including rolling out the alpha version by a certain deadline.

Because a good thought-out within a few man-days will allow you to think through and discuss all the points in advance. The fact that no one can do this is just human stupidity. It is necessary to develop more such a quality as perfectionism - it fundamentally will not allow something to be redone, because traces remain from the alterations, and this is no longer an ideal. Wouldn't that save more resources than it wasted?

If you are so smart, why are you working on this project, and not being its owner or at least a team leader who will listen to you exactly how everything should be done?
Or maybe your "perfectionism" lies in understanding a small part of the work that you do, and you have very little understanding about everything else that happens in the project?
For example, I, as a devops, would be very happy with the alpha version, so that I have something to work with. So that I could start preparing deployment automation, I looked at how everything starts in general. And do not wait for your development a few man-days, and then do everything in a hurry at night, because in the morning everything needs to start everywhere.