I asked this question:

Smallest/Simplest, modern pure relay computer with at least 4-bits data words

and found that it may be considered off topic in Retrocomputing. Another user advised me to ask here, and so I'm doing.

My intention is to build an new electromechanical computer (by new a mean a design made by me, but without reinventing the wheel), but restricting the technology used to purely electromechanical components.

One of the ideas behind it is to build a small thing without hundreds of parts, using only tech that can be found in some Zuse's machines, and in the Harvard Mark I and II computers. For example, in these last two, even the clock was electromechanical, being a shaft connected to an electric motor.

Any advise will be welcome. Thanks.

1 Answer 1


There is already a similar meta-question tackling this issue: Are questions about modern re-implementations on-topic?. It shows well that there is some reason to say yes, but as well as rejecting it.

In general, I do see your project as on topic, and here's where the missinterpretation hits. The rejection of your question was not about being on topic for the project, but for the way it was asked. The way it's made it asks for opinions and might even fall into the 'smallest / best / fastest / biggest / most awesome' cathegory of superlatives (somewhat related to this). Both is not well received either.

And that's, at the core, why I voted for close due 'calling for opinion'.

The question basicly asks for ideas to start with your project. While this is a great topic, it's none for RC, as it looks for opinions and even more suggestions about design. Eventually great for some free form forum, just not for RC.

To make it on topic it needs to be a specific and with relation to old machines. For example a question to understand Zuse's carry look ahead logic is perfect on topic, similar for details about his wireing technics, but asking for carry lookahead or soldering in general isn't

On a side note, I would also considere related questions about certain design decisions off-topic - like (stupid example) how to do an AND with switches - as they are about basic knowledge and in no way RC related. Just because something is done with parts that are no longer mainstream (for this purpose) doesn't make the question on topic. The right resources here are sites about basic logic - or eventually EE if it's about contact loads and alike. Last but not least, RC is neither a support forum for old computer software nor building new old hardware.

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .