Hacker News new | comments | show | ask | jobs | submit login
Bootstrap or not? High level pros and cons of using Twitter Bootstrap (baregit.com)
55 points by Loic 2075 days ago | hide | past | web | 32 comments | favorite



Bootstrap is a godsend for the corporate world and the vast majority of industry software shops that don't have huge levels public web exposure. Most of these places don't have the resources to hire dedicated UI specialists with the hottest CSS 3.0 skills, are still forced with IE6 somehow, and end up being designed by devs who liberally use nested tables and iframes like it's 1999.

I assume this dislike for bootstrap comes from those who believe it wants to compete as the de facto UI for anything of large scale or designed for instant sex appeal. Anything on that level will have their CSS highly customized and optimized by their specialized UI staff. I can also assure you, CSS and choice of UI framework is the absolute least of your priorities over business development, marketing, revenue generation, and security holes. Get the product out the door, and see if people actually want it before you worry about them inspecting your page source and find that you've used bootstrap.


I can also assure you, CSS and choice of UI framework is the absolute least of your priorities over business development, marketing, revenue generation, and security holes.

Unless, of course, people find the design of your MVP so utterly boring that they dislike the product entirely. I don't believe that design can be treated as an optional feature, that it's something you can add on later.


Here's what I find interesting:

Take a look at the computer you are reading this on. Mac, PC, Linux, other? Every single program on your machine has a uniform UI. Sure, there might be tweaks here and there and differences in icons, but the UI is consistent. You get on any Mac and things look and work the same way. You get on any PC (save Windows 8!) and everything looks and works the same way.

Has anyone ever said: I am not going to use <insert very useful software tool here> because the UI looks exactly the same as the UI on <insert another common software tool here>? Probably not.

Well, why is it that every site on the 'net has to look and work differently? If form follows function then most sites don't really need to reinvent the wheel. Yes, they need to deal with app-specific paradigms, but that's no different than the difference between, say, Excel, Word, Photoshop, iTunes and Dreamweaver.

Somehow on the desktop we have come to understand that a stable, uniform and well-understood UI is actually useful. While I am not sure that this translates directly into the web, I don't really agree with the idea that lots of sites using Bootstrap is equivalent to the end of the universe as we know it.

Considering how many downright awful sites there are this is probably an absolute gift from the Flying Spaghetti Monster (who's name shall never be uttered in vane).


"Every single program on your machine has a uniform UI"

Minesweeper, Internet Explorer, Windows Media Player, Windows Live Mail, Windows Live Messenger, Word 2010

6 Applications by a single vendor. 6 UIs that are similar (WIMP paradigm et al) but different (UI looks) - just like any 6 websites are similar (links, some forms, 2 or 3 pane layouts, ...) but different (design details).


Why are there so many of these debates?

This is really starting to be like php vs. ruby vs. dot net discussions. The end result is it doesn't matter as long as you're comfortable starting and finishing a project with it.

The value Bootstrap brings is to save time. Also, no one really should care what developers think of this as the teams race towards finishing their MVP. Your job is to finish the MVP - that's it!

IF YOUR PAYING CUSTOMERS are bitching at you for making yet another Bootstrap site, then let's come back and debate this. You guys are all focusing on the wrong problem.


I think most notable backlashes against Bootstrap came from the designer community. They do have valid points, but one cannot ignore the motives of the designers. Obviously due to Bootstrap, importance of designers at very early startup stage has diminished significantly.

This debate around Bootstrap seems really similar to enterprise web developers who used to use Java and C# complaining about Rails and Django a few years ago.


Reminds me of the think-geek t-shirt from days gone by "go away, or I will replace you with a very small shell script".

If all you brought to the table can be replaced with a single CSS file and a little bit of javascript maybe your value proposition wasn't really what you thought it was.


So... if it's a problem with designers not liking bootstraps stale design, why not build templates that use Bootstrap as a base and release them to the public?

I really don't see the issue.


Two plugs in a single thread is pushing it, but please see my other comment regarding Bootstrap themes/templates: http://news.ycombinator.com/item?id=3701287


Yesterday me and my girlfriend were looking for a car related service, and we found a site which supplied that service. It was obvious that it was using some kind of generic design template, so both of us got a bit suspicious, and not until we had researched the company for quite a while were we OK with doing business with them. It just didn't feel serious.

I feel the same way with a Bootstrap site that isn't customized at all - I would hesitate doing business with them, because if you don't have the resources to put a day on at least _customizing_ your design, are you really trustworthy? Of course you can be, and it's not a completely rational feeling, but I would be surprised if I am the only one with that gut reaction.

For minor tools (such as http://dochub.io) and other small projects, Twitter Bootstrap is awesome and doesn't even need much customization, but I don't think it's a good idea to use it uncustomized for a business.


I know several people on HN who have successful software (SaaS) businesses who will tell you that this just isn't really the case; that the overwhelming majority of your customers for something like car buying aren't going to finger you as a "Bootstrap user" and then think less of you for it.

It's hard not to wonder whether the fact that you're the kind of person who posts on HN doesn't bias your response to sites like this.

One way to get some real data on this would be to ask your friends to ask their nontechnical friends and relatives for a list of sites they actually paid money at, and then take a hard look at how good their designs really are.


Yesterday me and my girlfriend were looking for a car related service, and we found a site which supplied that service. It was obvious that it was using some kind of generic design template, so both of us got a bit suspicious, and not until we had researched the company for quite a while were we OK with doing business with them. It just didn't feel serious.

On the other hand you can find plenty of services/sites that looks graphically really professional and legit and that are instead plain scam (think about fake ecommerce sites)


The problem with Bootstrap is people don't treat it as a framework, they treat it as a theme. If people want to use Boostrap they can and should do custom themes that don't look like the tutorial. But they don't, they include the css/js, set some 'row', 'span4' and are happy with the result, because the navbar, forms and buttons are fancy startupish.

Simple conclusion: take advantage of the technical solutions Bootstrap provides and ffs put some work to make it look at least a bit original and branded.


For most people's applications, and in particular for the space of apps for which people pay money, it matters not- one- whit how much different your app looks than Bootstrap's native theme, because Bootstrap's native theme looks better than 80% of all web apps normal people come into contact with.

Different- for- different's- sake is a curse on web developers; it costs more time, more momentum, and more initiative than any other issue I can think of. Get the value working and proven first, then, when people are paying for it, start thinking of how much you'd budget to make your application visually distinctive.

There are exceptions, but they tend to prove the rule.


> Bootstrap's native theme looks better than 80% of all web apps normal people come into contact with

Today. The more people use raw TB the faster that 80% approaches zero and you'll get to my conclusion.


You're right; I left out the last part of my sentence, which should have read "... and yet still pay money to".

You will probably not pull off the next Foursquare or Pinterest using Twitter Bootstrap. But: you're not going to pull off the next Foursquare or Pinterest anyways.

So think of it this way: this debate has the question backwards. If you're doing a startup for which a major risk of your version 1 is that you built it with Bootstrap, you probably picked a bad startup idea.


Agreed. More sites need to move away from the default styles and should use Bootstrap for what its great at being -- a structural foundation.

I run a little marketplace for Bootstrap themes called WrapBootstrap - http://wrapbootstrap.com/ - I just sold the 100th theme today so more people are taking customization seriously within the Bootstrap community than before.

You can also check out Bootswatch - http://bootswatch.com/ - He provides free color themes that you can use to skin Bootstrap with.

People should definitely be using Bootstrap for its technical merit and not for its default styles.


I use Sass with Bootstrap without a problem. If you're the kind of person that has a hard time moving from one preprocessor to another, consider whether that's a VERY STRONG argument to go with something like Bootstrap, because it solves a lot of other problems that you could lose whole days or weeks to.


If you're using Sass with Bootstrap, then you're probably using the compiled version of Bootstrap (i.e. the final .css file.) If that's the case, you're missing out all the LESS mixins and variables that Bootstrap provides.

If you're serious about adopting Bootstrap, you should consider switching to LESS.

The less-rails-boostrap gem makes it easy if you're working on a Rails app. It integrates seamlessy with Rails' asset pipeline, and has a generator for Rails scaffold CSS.

https://github.com/metaskills/less-rails

https://github.com/metaskills/less-rails-bootstrap


You can also use the sass forks of bootstrap (700 watchers)

https://github.com/jlong/sass-twitter-bootstrap


That's awesome. Glad to hear a Sass port exists.

Out of curiosity, (and assuming you use the sass-twitter-boostrap gem in your projects) do you use any other Sass libraries besides Bootstrap port?


I use the Compass library which works great with Bootstrap. The recent Railscast also got me testing out the [Bourbon gem](https://github.com/thoughtbot/bourbon) (a similar Sass libary) on a new project along with Bootstrap.


Could you describe your workflow?


It might be a non-issue, but I am concerned about the performance of Bootstrap on older/slower computers as well as mobile devices. The stylesheet is so huge, I assume there has to be some kind of performance penalty somewhere? Or maybe I'm just overly paranoid... thoughts?


I wouldn't mind if the pendulum swings a bit towards unique functionality and away from unique aesthetics. A lot of the energy spent on unique look and feel could be better spent on usability and functionality.


"I wonder why Less and SASS are not merged in a single project"

I've also wondered this.


I think it is a given that you shouldn't use boostrap in it's entirety for a startup. However, using bootstrap for parts of the library is very effective. I'm using the excellent 'Bootstrap-sass-rails' gem, which lets you include all or just parts of bootstrap easily in the Rails asset pipeline. I then get the mixins, forms, type treatments, dropdowns, css reset and more for my project. Without the generic graphical styles.

Also, all of the less is converted to SASS.

Here's the gem if anyone is interested: https://github.com/yabawock/bootstrap-sass-rails


> Harder to be original. Because we all start with the same base.

That is so not true. What is easy is to use the basic theme to build out a prototype of your app. Customizing will always be harder, but then again, it's been hard to build out a well-designed site since the very beginning.

To say that it's hard because it comes with a default theme is to not understand.


The default theme makes it easier not to be original, but I don't see how it makes it harder to be original anywhere.


It doesn't. Sounds like there's a lack of imagination in that article or they're alluding to the inevitable result that people who possess neither the skills nor interest in making a visual customization of their site will leave it standing on the default theme, populating the web with a lot of sameness.

If it makes it easier for people to communicate faster, than I'm all for it.


I wrote the article, so please let me try to reformulate a bit better. When you start with a so complete, consistent and well though base design, it is hard to introduce new original concepts. This is because these new concepts will have to somehow fit nicely with the base design. It is kind of attracting you to simply do incremental improvements not to lose the benefits of the base design.

I am sorry, I am not a designer and not a native English speaker, this makes it a bit harder to formulate this correctly.


> I am not a designer

Then why would you even be talking about Bootstrap, let alone criticizing it from a design point of view?

You can take any past, present or future "original" concept and build it with Bootstrap. You do understand that, right? You are not limited to the default theme. If you're lazy, lack talent or interest then yes, your website will look pretty much the same as those of your peers, but then you can only blame yourself, not a CSS framework.




Guidelines | FAQ | Support | API | Security | Lists | Bookmarklet | DMCA | Apply to YC | Contact

Search: