I can confirm this. I got burned one time on a project cause I didn't know that. Had to scramble and make a working flexbox for it instead...which did okay fortunately. Pretty happy I did t get super complicated with it or I would've been boned.
That's the main reason why I've never used any Bootstrap JS.
EDIT: Several commenters have clearly not noticed that I said "Bootstrap JS", and not "Bootstrap". I still use Bootstrap (since v2), just not their Javascript.
I found React Bootstrap way more complicated than just using React and Bootstrap classes. React Bootstrap just adds more overhead to accomplish the same thing.
Yes. I don't want to pull in yet another library that is essentially web programming patterns from the year 2000 just to make a modal dialog go "Bing!" Vue is all I need.
I used Bootstrap 4 on a project recently, and thought I'd get away without jQuery - I almost did! The one place I needed it was for Bootstrap modal dialogs, to wire up open/close event handlers. I spent about an hour trying to figure out how to do it without jQuery, figured it was a deep rabbit hole, and caved...
Its funny how contentious modals are. Nearly every UX dev I know utterly despises them, yet I've seen at least two dozen UI implementations. Since I'm already neck-deep in VueJS, there are several easy ways to implement one, but now that I've seen bootstrap-vue, (several comments below), I may spend a few hours analyzing that over the holidays.
Lots of niceties in bootstrap-vue; I've been digging in to over the last several months. Biggest "drawback" is it's all very tied to vue2 (naturally). I know vue3 has been a moving target for a while, but it's out now. I think it'll probably be next summer before BSV is ported to vue3 (which, IIRC, will also be tied to bootstrap 5). Big undertaking, for sure.
I can't wait for them to break their "we'll maintain the previous version until the next version is stable".
Why would anyone believe Twitter would have a good release schedule with bootstrap 5 knowing how they has no stable supported version of bootstrap for a long time between 3 and 4.
Builtin RTL (Right-to-Left) support is great. I have used Bootstrap with different RTL flipping methods. however - recently I have switched to postcss-rtl ( https://github.com/vkalinichev/postcss-rtl ) which I find much better than other flipping solutions.
I sort of see where you come from, but it's practical, easy, and covers all the bases. Nobody ever got fired over choosing bootstrap, so to speak. It may not be a bold design choice, but for most, good enough is good enough.
tailwind is the new darling (for good reason IMO), and it can only see BS with binoculars.
It's like asking if any developers are still using Chrome. They are, overwhelmingly, even though I am convinced that Chrome is inferior, spyware and being used to strong-hand the internet. I wish I could change reality, but at least let us acknowledge it.
It is a trait of the news media to state their wish as established fact ;)
A lot of people try to knock down Bootstrap (just like they do jQuery and PHP) because it makes all websites look the same.
But Bootstrap has eased a lot of pain for a lot of developers like me and made a lot of bad websites look very good easily.
I greatly applaud everyone associated with this project esp Mark Otto and Jacob for this amazing thing they have given to the web and their non-stop efforts to make it better and better.
And imagine regular users being able to get used to a certain UX style instead of having to learn to navigate each new website differently. I gave up explaining my relatives how to use web beyond the 2 most important websites because each new website is a new adventure
For most of your target audience, this is a feature, not a bug. The exception is if you're targeting designers or web devs, of course.
People like seeing somewhat familiar designs. They know where to look for things. They know how to navigate. It provides a sense of trust because it looks vaguely like other professional brands they already trust.
Bootstrap has been amazing for accelerating web design across the industry. Sure, there are plenty of examples of better custom designs, but at what cost? Not all of us have extra budget to pay a team of designers and web devs to create the perfect custom design for our website.
100% agree. It's same with Windows apps that used to look the same and you instantly know how to navigate them. Now the web is mishmash of idiotic solutions with videos behind text, scrolling with no indications whether it's left- right, or top-bottom. I understand people try to showcase their skills sometimes, but good design, most of the time, is 'less is more'.
One great thing about bootstrap is the availability of good quality themes.
They don’t really make websites based on them entirely not look like bootstrap sites, but there’s a decent array of choices and variation for many kinds of projects.
Agreed. Bootstrap is my default. I’m already on 5 for some as-yet-unrelesed projects, and it’s great.
My major criticism in the past was the difficulty of migration, but that seems to be better addressed in this version. (I have not yet tried to migrate, though).
What I love most about Bootstrap is its extensibility, and especially its decomposability.
It's not widely appreciated that you can take subsets of Bootstrap, individual components for example, or its typography system, to use in non-Bootstrap scenarios.
This is an acid test this project aces, easily. Bootstrap is the first place I look when I'm missing something in another project.
Maybe I'm being picky but I'm a bit bummed not to see some sort of dark mode support here, especially after Tailwind released fantastic dark mode functionality last month.
I apologize beforehand if this is spammy or self-promoting, but I am building a framework which is basically an alternative to Bootstrap with a dark mode: https://www.gethalfmoon.com/
That's definitely a good thing, but I think the person you're replying to is also concerned about the performance of the code when being run too. Things like dom manipulation could be slower in BS5's new JS, while jquery's has likely been heavily optimized over the years.
Off-topic, but I clicked this link, watched the Jackson 5 video, got lost on Wikipedia, and closed the tab before reading the article (which I was initially excited to read).
The problem with the internet isn't clickbait and hyperbole, it's that it's changed our attention spans.
Regarding attention spans, from "Avoid News. Towards a Healthy News Diet" By Rolf Dobelli [0] (top-notch article, btw):
"In a 2001 study [1] two scholars in Canada showed that comprehension declines as the number of hyperlinks in a document increase. Why? Because whenever a link appears, your brain has to at least make the choice not to click, which in itself is distracting."
This had me thinking if a dead simple button at the end of an article "Enable links" would have a positive effect on link heavy articles (Wikipedia type) or not. Something like a conscious action of "okay, now I'm ready to explore this further". Or, better yet, a browser add-on that will detect just the content of an article and disable its links by default. Obviously, with an enable button close by.
Another problem with the Internet is that it’s made everyone really excited about getting on a soapbox to make sweeping and categorical generalizations about our zeitgeist based on nothing but highly personal experiences.
I'm still using Bootstrap and Jquery for my side projects.
It helps lower the amount of technology I use in my stack.
From my experience maintaining frontend technology is not as simple as backend.
I've come to the conclusion that data is the most important thing, so my concentration is mostly on getting the database right then build out with a backend rendering framework. So the presentation is not as important in the initial phases and bootstrap is good enough.
I've used bootstrap in the past as a non-designer to get a decent set of defaults for web applications. I definitely appreciate that aspect. But I am wondering if there is a lighter weight alternative. Maybe using reboot directly?
So many of the reasons I loved Bootstrap revolved around the limitations of CSS 7-10 years ago. CSS Grids, Flexbox, and browser standardization has pretty much eliminated my desire for a CSS framework.
Loved Bootstrap back in the day. Not sure I see the point anymore though.
How thoroughly do you test all of your components? For example, do you always test keyboard navigation when making changes to the styling of your components that might conflict with keyboard navigation?
The reason I'm asking is because I've used or evaluated dozens of CSS frameworks and React component libraries, including custom solutions developed by the companies I've worked for, and none of these solutions are as well-tested at Bootstrap. And one example of the issues I've encountered with these other solutions is bugs relating to keyboard navigation (e.g., Ant Design and Semantic UI were two of the worst offenders the last time I evaluated them).
For many developers nowadays who grew up with touch-based UIs and can't touch type, keyboard navigation might seem like an odd thing to focus on. But I've found that checking if CSS frameworks and React component libraries fully support keyboard navigation is a quick way to evaluate them, like how David Lee Roth evaluated concert promoters by asking them to remove brown M&Ms.
And if the teams that made Ant Design, Semantic UI, etc. can't adequately test their components, how capable is the average developer of creating their own well-tested CSS framework?
In general, my philosophy on components is: Don't reinvent the wheel.
Whenever possible, I try to extend stock components rather than create something new. If you use a select instead of making some react-based div-popper thing, accessibility is included. It can be tougher to tweak them to look the way you want, but you don't have to worry as much about what a screen-reader is going to do with it, etc.
Totally agree about how bad some sites are with regards to accessibility.
67 comments
[ 3.0 ms ] story [ 99.7 ms ] threadhttps://getbootstrap.com/docs/5.0/getting-started/browsers-d...
Also Bootstrap 3 to 4 have had patterns of dropping support of IE versions.
EDIT: Several commenters have clearly not noticed that I said "Bootstrap JS", and not "Bootstrap". I still use Bootstrap (since v2), just not their Javascript.
https://react-bootstrap.github.io/
https://bootstrap-vue.org/
data-bs-toggle="dropdown"
That's some ... cuteness ;)
Why would anyone believe Twitter would have a good release schedule with bootstrap 5 knowing how they has no stable supported version of bootstrap for a long time between 3 and 4.
https://getbootstrap.com/docs/4.5/about/overview/
tailwind is the new darling (for good reason IMO), and it can only see BS with binoculars.
It's like asking if any developers are still using Chrome. They are, overwhelmingly, even though I am convinced that Chrome is inferior, spyware and being used to strong-hand the internet. I wish I could change reality, but at least let us acknowledge it.
It is a trait of the news media to state their wish as established fact ;)
But Bootstrap has eased a lot of pain for a lot of developers like me and made a lot of bad websites look very good easily.
I greatly applaud everyone associated with this project esp Mark Otto and Jacob for this amazing thing they have given to the web and their non-stop efforts to make it better and better.
You can easily override the defaults with SCSS variables and customize it to look however you want.
Most sites look similar now anyway.
For most of your target audience, this is a feature, not a bug. The exception is if you're targeting designers or web devs, of course.
People like seeing somewhat familiar designs. They know where to look for things. They know how to navigate. It provides a sense of trust because it looks vaguely like other professional brands they already trust.
Bootstrap has been amazing for accelerating web design across the industry. Sure, there are plenty of examples of better custom designs, but at what cost? Not all of us have extra budget to pay a team of designers and web devs to create the perfect custom design for our website.
I hate those stupid apps that use custom interfaces and custom buttons and whatnot. And for what? "Branding"?
Somehow f.lux gets a pass from me here but it's on thin ice ;)
I used to prefer Zurb‘s foundation because I liked their defaults better, specially regarding tyopgraphy, but boy they ruined that framework.
They don’t really make websites based on them entirely not look like bootstrap sites, but there’s a decent array of choices and variation for many kinds of projects.
My major criticism in the past was the difficulty of migration, but that seems to be better addressed in this version. (I have not yet tried to migrate, though).
Keep up the great work Bootstrap team!
It's not widely appreciated that you can take subsets of Bootstrap, individual components for example, or its typography system, to use in non-Bootstrap scenarios.
This is an acid test this project aces, easily. Bootstrap is the first place I look when I'm missing something in another project.
Bootstrap 4.5 starter kit [1]: 74kb, DOMContentLoaded after 180 ms
Bootstrap 5.0 starter kit [2]: 57kb, DOMContentLoaded after 160 ms
[1] https://getbootstrap.com/docs/4.5/examples/starter-template/
[2] https://getbootstrap.com/docs/5.0/examples/starter-template/
The problem with the internet isn't clickbait and hyperbole, it's that it's changed our attention spans.
"In a 2001 study [1] two scholars in Canada showed that comprehension declines as the number of hyperlinks in a document increase. Why? Because whenever a link appears, your brain has to at least make the choice not to click, which in itself is distracting."
This had me thinking if a dead simple button at the end of an article "Enable links" would have a positive effect on link heavy articles (Wikipedia type) or not. Something like a conscious action of "okay, now I'm ready to explore this further". Or, better yet, a browser add-on that will detect just the content of an article and disable its links by default. Obviously, with an enable button close by.
---
[0] https://www.gwern.net/docs/culture/2010-dobelli.pdf
[1] Nicholas Carr: The Web Shatters Focus, Rewires Brains, Wired, May 2010
Fair point, though.
It helps lower the amount of technology I use in my stack.
From my experience maintaining frontend technology is not as simple as backend.
I've come to the conclusion that data is the most important thing, so my concentration is mostly on getting the database right then build out with a backend rendering framework. So the presentation is not as important in the initial phases and bootstrap is good enough.
Spoiler alert, there rarely is a need.
- Bootstrap - Jquery - Handlebars
should be enough for many small to medium sized applications
Loved Bootstrap back in the day. Not sure I see the point anymore though.
That's interesting. What do you use in your projects now? Tailwind? Or maybe just plain old CSS?
The reason I'm asking is because I've used or evaluated dozens of CSS frameworks and React component libraries, including custom solutions developed by the companies I've worked for, and none of these solutions are as well-tested at Bootstrap. And one example of the issues I've encountered with these other solutions is bugs relating to keyboard navigation (e.g., Ant Design and Semantic UI were two of the worst offenders the last time I evaluated them).
For many developers nowadays who grew up with touch-based UIs and can't touch type, keyboard navigation might seem like an odd thing to focus on. But I've found that checking if CSS frameworks and React component libraries fully support keyboard navigation is a quick way to evaluate them, like how David Lee Roth evaluated concert promoters by asking them to remove brown M&Ms.
And if the teams that made Ant Design, Semantic UI, etc. can't adequately test their components, how capable is the average developer of creating their own well-tested CSS framework?
In general, my philosophy on components is: Don't reinvent the wheel.
Whenever possible, I try to extend stock components rather than create something new. If you use a select instead of making some react-based div-popper thing, accessibility is included. It can be tougher to tweak them to look the way you want, but you don't have to worry as much about what a screen-reader is going to do with it, etc.
Totally agree about how bad some sites are with regards to accessibility.
https://news.ycombinator.com/item?id=25060568