Bootstrap, We Hardly Knew Ye
Well that was kind of an aside, but actually on topic of what I want to talk about here. I got back into Bootstrap and thought "Wow, this is awesome," but then one of the guys I was working with pointed out that most of the cheap little Bootstrap templates don‘t come with the LESS files. I thought "Oh Jesus, here we go." Just like eclipse, it‘s free, but then it just becomes a big baby jumper with a bajillion barnacles trying to cop a ride. Now there‘s also the whole thing that most templates are built for Bootstrap 2 and there are none for 3 yet. Then, there‘s the matter that once you have applied your little template, while your creation is certainly much more malleable and extensible, it‘s also kind of stupidly narrow in what it is. Try to combine two templates some time. It makes blinders and a feedbag seem like a noble fate.
Of course, you can ignore templates and just develop your own pages. Bootstrap does give you a good start if that‘s your plan, and there are two tools that are worth considering: Jetstrap and Divshot. I have to say I didn't like the pricing schemes of these companies, and I found that their vision is not really clear. So Bootstrap got us something to start with but then we just started making the pages, and then what? Jetstrap‘s demos make it seem like they are mostly focused on content creation, e.g., just getting the text in place.
None of these tools talks about integration with programming frameworks.
Anyway, the other day after some work with Bootstrap, I saw a tweet about Webflow. I think I had heard of these guys and thought "Why would I do that?" But then, I went in with SQURL and scooped up their videos and watched them and found that I was quickly impressed. On the plus side of this, you watch these and start to think really quickly how RIDICULOUS it is that we have had almost two decades of the web and nothing like this has existed. It makes the idea of paying people $125 an hour to navigate the Adobe bloatware look like plowing a field with a pencil.
Here‘s a really stunning part of this project: it‘s the work of three people. What does this say about open source? For some reason, OS is UI-constipated. We know that. Bootstrap went from a two-person ditty to the biggest project on GitHub last year, and this app makes those guys look like tin-pounding primates. Sure, Bootstrap is open, but what does that really end up buying you?
Here's the bigger circle I want to draw around this, the one I mentioned the other day in relation to Mongo: I see a future in which the little tweaker types are largely just the pheromone following ants who find the juicy piles and then large artillery is rolled in behind. I don‘t know that this thing (Webflow) is really large artillery, but it is responsive and interactive and it might just be fodder for another tool.
While this has been going on I have been using IntelliJ some more on my Play project. As just a straight-out IDE, it beats Eclipse in almost all categories. The template mechanism is greatly superior. Many of the things I have complained about in Eclipse for years are easy in IDEA, e.g. capitalizing a variable. I made two templates for making Bloch-style static builder implementations that are fantastic. The main problem with IDEA is that it is so painfully slow I feel like I‘m trying to pass a meal that had ground glass in it. Running a single unit test in it takes FOREVER! Eclipse is faster, even though the M2 plugin fixes slowed-down test runs. Lately, I've just been doing single runs from the Play console. Xcode is so much faster at running tests, it's laughable. And there really is no solution to this: waiting around thinking that UI Java apps are finally going to become efficient is Tooth Fairy grade thinking at this point. What we need is someone to look at this product as representing just the state of the art in terms of consensus about what the IDE should do, and come in and rewrite this damned thing in C++ or something like that. Maybe JetBrains will wake up and realize the need for that at some point. I doubt it. If a quad core with 8GBs of RAM and an SSD is not fast enough to get your dancing hippo off the ground, you should really think about a complete rewrite.
Finally, consider how ludicrous it is that the sole focus of development in the last decade has been to make it so the poor programmer has to type less. The stupidest Ruby argument about how Java is verbose could be done 10,000 times in a given day just while you are waiting for tests to run. Which is hilarious, because it makes you realize that the developer delivers the "more productive" argument, but is really saying "Owie, Java makes my fingers hurt ..." even though IDEA will do the verbose crap for you, and even fold it when it‘s done so you don‘t have to look at it (e.g., List items = new ArrayList()). This is kind of the definition of bread and circus, as far as I can tell: the developers are the mob and the language/tool millers are the regime, trying to keep themselves from a bloody mauling.
I had a brief exchange with one of the founders of Webflow. I said that it seemed like their focus was only on designers, and he said they were starting with the goal of speeding the PSD-based cycle of doing static stuff, then they wanted to do battle with Wordpress. Um, Bootstrap already kicked that to the curb. Well, at least the idea that we need to make sites from a blogging tool because blogs require new content. Now that I have thought about this though, what we really need is an app that can sit in between the UI builders and the templating tools. I know going back and forth between template languages is horrendously complex, but given the separation that‘s now possible with LESS/CSS, it's probably worth revisiting ...