Has anyone who is doing Web Development had a look at Google Polymer and/or Mozilla X-Tag web components for their company or client's website designs? Or is it more the case where sites don't want to change just yet (or at all)?
Now I need to work out how to get it running on the internet, not just on the development server on my laptop. All the re-writing and creating new features and learning Django and all that has been really good fun. This next step is going to be HELL!!!
AAAAArrrrhghghhhgh!!! How can everything be just so obtuse? Three attempts later, on three different days, and all I've managed to do is break and fix and break and fix the old static html version of the website a few times. The django version is still just hanging out on my own laptop, no closer to making its way onto my server. Maybe next time I'll get somewhere, anywhere. But I've a feeling that past mistakes are now somehow baked in, and I'm not sure how serious it might be to just delete everything and start over. Soooooo annoying.
Now I need to work out how to get it running on the internet, not just on the development server on my laptop. All the re-writing and creating new features and learning Django and all that has been really good fun. This next step is going to be HELL!!!
AAAAArrrrhghghhhgh!!! How can everything be just so obtuse? Three attempts later, on three different days, and all I've managed to do is break and fix and break and fix the old static html version of the website a few times. The django version is still just hanging out on my own laptop, no closer to making its way onto my server. Maybe next time I'll get somewhere, anywhere. But I've a feeling that past mistakes are now somehow baked in, and I'm not sure how serious it might be to just delete everything and start over. Soooooo annoying.
Yes. It's hard. I've written a tutorial on it, but it's slightly outdated.
Not sure what to do next... but this is progress of sorts.
Are you using the manage.py runserver? You can't use that in production, it's for local development only. You have to use something else like uWSGI or Gunicorn.
Oh no, don't worry about that. I'm going to be using Passenger wsgi, which is a simple tick-box on Dreamhost. I've just not got it to work yet. Because, of course, it's not just a simple tick-box. Of course.
Oh no, don't worry about that. I'm going to be using Passenger wsgi, which is a simple tick-box on Dreamhost. I've just not got it to work yet. Because, of course, it's not just a simple tick-box. Of course.
Well, good luck with that. In theory managed hosting should make things easier since you have less to learn and someone else is taking care of the hard parts. In reality it causes troubles because you have a lot less control over important bits. Also, most of the documentation out there is written assuming you are using plain old *nix servers, and it's much harder to get help with problems on managed hosting because the knowledgeable people to ask don't use it. If you really want to use a managed hosting solution, Heroku might be better than Dreamhost.
Oh no, don't worry about that. I'm going to be using Passenger wsgi, which is a simple tick-box on Dreamhost. I've just not got it to work yet. Because, of course, it's not just a simple tick-box. Of course.
Turns out it is working. Any link to anything not in the public folder gets picked up by Passenger, and displays that page. That isn't the development server, it's just the default page for an empty Django project. Which is cool, I guess. At least I can get the new site working in my own time and the old site will just remain as long as needed.
Now I'm stuck with trying to get the script that populated the local sqlite database on my laptop to do the same thing with the MySQL database on my server. Or to synch the databases. And again, I simply don't know enough about this stuff to get it sorted without my brain exploding for a few days in a row.
To simultaneous turn or not? That is the question.
I say thee nay. Simultaneous introduces too many situations we don't have answers for. The only flaw in turn taking is there may be an advantage to the first or second mover depending on the circumstances.
One solution is fog of war. Not being able to see your opponent's earlier moves makes it really not matter very much who moved first since you both get the same info whenever you meet.
Another solution is to take turns making one decision at a time. I really like this idea. It eliminates the problem where one player takes a nap while the other player takes their turn, maintaining player engagement at all times. It makes it really not matter very much at all which player started the game. It also makes battles feel more like battles where both sides are moving and shooting. In Advance Wars often times one player can completely rout their opponent on a single super CO-powered turn without giving them a chance to even do anything but die.
To simultaneous turn or not? That is the question.
I say thee nay. Simultaneous introduces too many situations we don't have answers for. The only flaw in turn taking is there may be an advantage to the first or second mover depending on the circumstances.
One solution is fog of war. Not being able to see your opponent's earlier moves makes it really not matter very much who moved first since you both get the same info whenever you meet.
Another solution is to take turns making one decision at a time. I really like this idea. It eliminates the problem where one player takes a nap while the other player takes their turn, maintaining player engagement at all times. It makes it really not matter very much at all which player started the game. It also makes battles feel more like battles where both sides are moving and shooting. In Advance Wars often times one player can completely rout their opponent on a single super CO-powered turn without giving them a chance to even do anything but die.
Yes, this is a pretty good analysis. One thing I would really like to avoid is making the game feel "puzzle-y" which is a symptom many tactics games can feel if the correct strategy is too apparent, i.e. providing too much information at a time. Simultaneous turns helps avoid this, as well as alternating moves.
I want the game to be reminiscent of tabletop wargames but without all the grognard bullshit.
One idea I've thought of that may or may not work. Take some ridiculous Grognard game, like Advanced Squad Leader. Now convert all the games rules into source code. So all that ridiculous simulation grognard stuff actually happens under the hood.
Now, create a user interface that hides almost everything from the user. Make the UI as simple as Advance Wars. Remove options from the player until they are left with something only slightly more complicated than moving and shooting. All that Grognard stuff is happening, but only the result is visible to the player. Without inspecting RAM, hardly any numbers will be visible. Reduce the possible inputs without reducing outputs.
This should result in something that feels alive and simulation-like. In theory you would have the simplicity of an Advance Wars with the emergence of a Dwarf Fortress.
When you attack sometimes you'll just lose despite the odds because the other unit was just full of badasses. Sometimes your vehicles will just fail due to poor maintenance or sabotage. Just all this crazy shit will happen and make the battlefield feel alive even though all you're doing is telling people to move and shoot.
That's pretty much exactly what I'm aiming to do. Really pair down the interface, make it super minimal and modern, but on the backend run crazy advanced warfare simulations. My goal is to have almost no numbers in the interface.
For each unit you get a couple of decisions like entrench, assault, cover, or specialty actions like call artillery or throw smoke. Also morale systems that automatically make your units do semi-random actions will also lead to the DF like experience.
If you actually finish it, that kind of game can really get a cult community forming around it. Ideally the forums will be filled with tales of epic times. The potential for this was seen a bit back in the days of Wesnoth during the massacre of Foxton.
I'd play the old Advance Wars and then Days of Ruin all the way through again for insight.
The former are all very puzzle-feeling. The latter feels very high level military/tactical. The "puzzles" are solved by "focus the air units on the south bases while the infantry plot through the mountains" instead of old AW games' "move the transport copter 4 spaces left and drop the infantry up, then move it two down and one left and move the infantry two up and one left....".
If the three of us had a brainstorming session on these games, we could record it and BAM! Podcast!
Problem is we did that level of effort for GeekNights. There's no way to do that AND something else significant full time unless we bail on real jobs and money.
That's pretty much exactly what I'm aiming to do. Really pair down the interface, make it super minimal and modern, but on the backend run crazy advanced warfare simulations. My goal is to have almost no numbers in the interface.
For each unit you get a couple of decisions like entrench, assault, cover, or specialty actions like call artillery or throw smoke. Also morale systems that automatically make your units do semi-random actions will also lead to the DF like experience.
Sounds a bit like extending Advanced Warfare with the feel of XCOM missions.
A possible alternate to simultaneous turns without wasting time on a turn by turn basis is having a variable for the time it takes a player to execute their turn which may influence critical chance or missing. However that may just lead to who can press buttons fastest (unless minimum grace time is present).
Comments
Or is it more the case where sites don't want to change just yet (or at all)?
Unity is pretty awesome.
Not sure what to do next... but this is progress of sorts.
It's the same as regular Advance Wars except the only units are mechs.
Fuck designing. Make it.
Now I'm stuck with trying to get the script that populated the local sqlite database on my laptop to do the same thing with the MySQL database on my server. Or to synch the databases. And again, I simply don't know enough about this stuff to get it sorted without my brain exploding for a few days in a row.
One solution is fog of war. Not being able to see your opponent's earlier moves makes it really not matter very much who moved first since you both get the same info whenever you meet.
Another solution is to take turns making one decision at a time. I really like this idea. It eliminates the problem where one player takes a nap while the other player takes their turn, maintaining player engagement at all times. It makes it really not matter very much at all which player started the game. It also makes battles feel more like battles where both sides are moving and shooting. In Advance Wars often times one player can completely rout their opponent on a single super CO-powered turn without giving them a chance to even do anything but die.
I want the game to be reminiscent of tabletop wargames but without all the grognard bullshit.
Now, create a user interface that hides almost everything from the user. Make the UI as simple as Advance Wars. Remove options from the player until they are left with something only slightly more complicated than moving and shooting. All that Grognard stuff is happening, but only the result is visible to the player. Without inspecting RAM, hardly any numbers will be visible. Reduce the possible inputs without reducing outputs.
This should result in something that feels alive and simulation-like. In theory you would have the simplicity of an Advance Wars with the emergence of a Dwarf Fortress.
When you attack sometimes you'll just lose despite the odds because the other unit was just full of badasses. Sometimes your vehicles will just fail due to poor maintenance or sabotage. Just all this crazy shit will happen and make the battlefield feel alive even though all you're doing is telling people to move and shoot.
For each unit you get a couple of decisions like entrench, assault, cover, or specialty actions like call artillery or throw smoke. Also morale systems that automatically make your units do semi-random actions will also lead to the DF like experience.
Bloody, Bloody Foxton.
The former are all very puzzle-feeling. The latter feels very high level military/tactical. The "puzzles" are solved by "focus the air units on the south bases while the infantry plot through the mountains" instead of old AW games' "move the transport copter 4 spaces left and drop the infantry up, then move it two down and one left and move the infantry two up and one left....".
If the three of us had a brainstorming session on these games, we could record it and BAM! Podcast!
EDIT: Almost forgot Dynasty Warriors.
A possible alternate to simultaneous turns without wasting time on a turn by turn basis is having a variable for the time it takes a player to execute their turn which may influence critical chance or missing.
However that may just lead to who can press buttons fastest (unless minimum grace time is present).