-
Notifications
You must be signed in to change notification settings - Fork 70
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Combine Synth with another project #52
Comments
Can anyone suggest a concrete way to roll Synth support into another project? If so, which project? I'll continue to modularize components of Synth so it can be used by other projects, but so far I don't really see how to do what's being asked in this issue. |
Hi Jon, I had read your message from #51 on my phone and didn't see your email address. So my comment above probably looked weird. Anyhow, this question of how to do it is a technical one that I, as a beginner, can't answer. But I have a collection of all the contact info from those who can. I think the best thing to do is for all of you to confer about this by Google Hangout or some other way. I'm missing a couple of them. I'd like to have Brad Green and Miško Hevery in the discussion too. I don't have an email address for Eddie Monge and Brian Ford from generator-angular either. I am happy to collect facts and create the case for joining forces, but nobody knows me, so it doesn't really matter what I say. I sent out one message so far to 20 addresses I had found looking for some input on my facts collection and only got replies from the rails-api guys who wanted to be dropped from the list. The technical effort has to follow a political one, which I think is going to be harder. It's great that you're at least open to the discussion. If you have any pull with anyone with some clout, getting their interest would be the best next step in my opinion. How to do it is more like step 10 or 20. |
Thanks for the suggestions Dan, but I don't have the time right now, nor the vision, to combine Synth with some other project. I'll keep my options open though if a good opportunity arises though. |
Linux succeeded because it had a dictator. Plenty of people wanted to run Unix on pcs. Without a dictator, we would have had a hundred failed, inadequate and competing attempts to do it and the world would be a much worse place. This new situation with app starters doesn't have a dictator. That's not a bad thing. Who likes dictatorship anyway? All it takes is dialog. Just talk to each other, understand the humungous gains of cooperation and you will find a way to agree. As far as I know, nobody has even tried talking to each other yet. |
Hi Jon,
I don't have your personal contact info so I'm doing this this way.
Synth has some cool features. The trouble is that when someone chooses to use it for their project they miss out on the different cool features of alternatives. Unfortunately, they can only use one thing to start their project.
As far as I can tell, the top cool features of Synth are
I think Preloading data could be added to another project. It might be the case that the second thing, Working with any JS framework, can't.
So, is the value of that benefit more than the value of what you miss out by not using any of the alternatives?
To help answer that question, I made this spreadsheet. It has separate sheets for comparisons between Synth and the top other generators. There's also a sheet with my thoughts about why we need to reduce the number of alternatives.
The spreadsheet attempts to give values to each benefit, so the whole generators can be compared. The lists of what benefits each one has are objective, but the importance weights are naturally subjective. Take them for what they are worth.
It makes sense for Synth to exist as a distinct generator under two conditions:
If (2) is never the case, then it will only make sense to keep Synth as a separate project if enough benefit is added to it to make it the case. That seems like it would be a lot of work, but I don't know.
So two questions:
To me, the idea of adding JS framework flexibility to the richest generator makes more sense in the same way that inheritance in object oriented programming makes sense - you only need to make the differences and you get what's common for free. What do you think?
The text was updated successfully, but these errors were encountered: