there should be a page that has brand guidelines for the logo.

• Logo.png
• logo.svg
• typography
• colors

Capital B "Bower"

Hi,

I've register a bower package aaa-settings-dialog. After registering the package I noticed that some of the metadata in the bower.json manifest was incorrect. I have corrected the local bower.json file and I now want the bower registry updated. How do I do this?

I've tried rerunning the bower register command with the same package name but this fails as follows:

bower register git://github.com/Will-in-BC/aaa-settings-dialog.git
bower EUNKNOWN Unknown error: 403

simple-feature-detector can't be found with "simple-feature-detector", "simple-feature", just "detector", etc. bower search detector returns it in the results.

bower/spec#2

There is no template for translate pages to other languages.this can help bower community.

Heads up! I'll be moving this repo to bower/bower.io in the next couple weeks.

@rayshan Let me know which UA id we should use.

We should provide a media kit with the logo in SVG and PNG and a short guideline how it should be used.

https://gist.github.com/desandro/1c50118441f703f3f6e1#file-bower-logo-svg

@desandro

The final bullet point in the first section of https://github.com/bower/bower.github.io/blob/master/docs/creating-packages.md has a typo.

Currently:
Set to true if you want to keep the package private and do not want to register the package in future.

Fixed:
Set to true if you want to keep the package private and do not want to register the package in _the_ future.

Consider using Jekyll or deploying elsewhere, once the site is a bit more complex. Can do what HTML5 Boilerplate does and have a post-push hook to deploy the site elsewhere. Jekyll would be simpler.

They should also follow the naming standards and not include version numbers, etc.

Pages are getting long...

Title suggestion: Use in production / Production use / DevOps?

Content:

• Move CI section from API
• CLI --quiet flag
• Analytics
• Running a private registry (docker / db replication / etc.)

Relevant issues:
bower/bower#1455
bower/registry#42
bower/registry#73 (comment)

cc @benschwarz

from the docs:
main - string or array: The primary endpoints of your package.

What does that supposed to mean?? what is an "endpoint" anyway? this is the worst description ever. Lets say I have a jQuery plugin, it has javascript, css and an image. what now? This explains nothing.

The bower channel on freenode could be a great place for users and component publishers to get help, but it's not very accessible to many potential users; the only mention is way down the page and not everyone is on IRC anyway. It would be far more accessible if there was a non-sucky live chat embedded on the page, something like this: http://scrollback.io/pwn/bower/bower.io

Disclaimer: I'm one of the guys that wrote Scrollback.

Additional notes:

• It won't connect people to the channel until they say something (no redundant join/parts)
• The server has an I-Line at freenode (hundreds of users can connect)
• It reports IP addresses to freenode so chanops can ban individual users.
• Infinite scrollback (duh) so people can refer to past conversations. Search is coming soon.

I can send a pull request if you're interested.

New logo: http://i.imgur.com/QmGHPgc.png

Update the website since bower README changed on 1.0.0.

Write detailed error report (number row and character)

I have:
Additional error details:
Unexpected token n

and can't fix

All four links in this section are relative links to CONTRIBUTING.md which become http://bower.io/CONTRIBUTING.md on this site.

Either use absolute links in the README or fix relative paths after import.

Invisible on the homepage, but visible on http://bower.io/search/, the current bower png version contains a dotted line along the right side which I suspect shouldn't be there. Amazing logo by the way! I was recognizing the logo online before I even knew the name!

Here's the import from gdocs:

The Bower configuration alters the behaviour of Bower on the local system.

The config is obtained by merging multiple configurations by this order of 
importance:

* [cli args via --config.*]
* environment variables
* local bowerrc located in the cwd
* all bowerrc files upwards the directory tree
* bowerrc file located in the home folder
* bowerrc file located in the global folder

Example of valid environment variables:

* `bower_endpoint_parser` is evaluated as `endpoint-parser`
* `bower_storage__cache` is evaluated as `storage.cache`

The bowerrc file mentioned must be a JSON valid file.

## cwd [string]

The directory from which bower should run. All relative paths will be calculated 
according to this setting.  

## directory [string]

The path in which installed components should be saved. If not specified this 
defaults to `bower_components`.  

## registry [object string]

The registry config. Can be an object or a string. If a string is used, all the 
property values bellow will have its value. Defaults to the bower registry URL.

If your organization wishes to maintain a private registry, you may change the 
values below.

## registry.search [array string]

An array of URLs pointing to read-only Bower registries. A string means only 
one. When looking into the registry for an endpoint, Bower will query these 
registries by the specified order.

**registry.register [string]**

The URL to use when registering packages. 

**registry.publish [string]**

The URL to use when publish packages.  


## shorthand-resolver [string]

Define a custom template for shorthand package names.  
Defaults to git://github.com/{{owner}}/{{package}}.git

The `shorthand-resolver` key provides support for defining a custom template 
which Bower uses when constructing a URL for a given shorthand. For example, if 
a shorthand of `twitter/flight` or `twitter/flight#v1.0.0` is specified in the 
package's manifest dependencies, the following data can be referenced from 
within the `shorthand-resolver` template:

owner: "twitter"  
package: "flight"  
shorthand: "twitter/flight"

Examples:

    "shorthand-resolver": "git://example.com/{{owner}}/components/{{package}}.git"
    "shorthand-resolver": "git://example.com/{{shorthand}}.git"

**proxy [string]**

The proxy to use for http requests.

**https-proxy [string]**

The proxy to use for https requests.

**user-agent [string]**

Sets the User-Agent for each request made.  
Defaults to: node/{process.version} {process.platform} {process.arch}

**timeout [****number****]**

The timeout to be used when making requests in milliseconds, defaults to 
60000ms.

**strict-ssl [boolean]**

Whether or not to do SSL key validation when making requests via https.

**ca [object string****]**

The CA certificates to be used, defaults to null. This is similar to the 
registry key, specifying each CA to use for each registry endpoint.

The Certificate Authority signing certificate that is trusted for SSL 
connections to the registry.  
Set to null to only allow "known" registrars, or to a specific CA cert to trust 
only that specific signing authority.

**color [boolean]**

Enable or disable use of colors in the CLI output. Defaults to true.

**storage**** [object]**

Where to store persistent data, such as cache, needed by bower. Defaults to 
paths that suite the OS/platform. Valid keys are `cache`, `registry`, `links`, 
`completion`.

**tmp [string]**

Where to store temporary files and folders. Defaults to the system temporary 
directory suffixed with /bower.

**interactive ****[boolean]**

Makes bower interactive, prompting whenever necessary. Defaults to `null` which 
means `auto`.

Chrome 27/Chromium 25

Edit: Silly Dropbox link expired

Right now http://twitter.github.com/bower/ → 404

$ curl -LIsS http://twitter.github.com/bower/

HTTP/1.1 301 Moved Permanently
Server: GitHub.com
Date: Wed, 01 May 2013 12:20:58 GMT
Content-Type: text/html
Content-Length: 178
Connection: keep-alive
Location: http://twitter.github.io/bower/
Expires: Wed, 01 May 2013 12:30:58 GMT
Cache-Control: max-age=600
Vary: Accept-Encoding

HTTP/1.1 404 Not Found
Server: GitHub.com
Date: Wed, 01 May 2013 12:20:58 GMT
Content-Type: text/html
Content-Length: 8456
Connection: keep-alive

I've seen a lot of links around the web pointing to http://twitter.github.com/bower/ (including in the bower-server repository description) so, I think this should be fixed somehow, either by a redirect or at least an empty repository telling people about the move (e.g.: paulirish/html5-boilerplate).

(Ported from bower/bower#354)

As people start to write tools to work with Bower in different development setups, we should maintain a list of the best / most useful tools. Hopefully it will encourage people to use Bower in a setting that they feel comfortable with. This would be something to include on the project page / website. For now, we can keep a catalogue in this issue:

alfred-workflow
A collection of Alfred workflows that includes Bower integration.

bower
Bower integration for your Ruby apps (sprockets).

bower-rails
rake tasks for Bower on Rails.

cakephp-bower
CakePHP Plugin for Bower

grunt-bower-concat
Grunt task for automatically concat all installed Bower components.

grunt-bower-install
Inject your Bower components right into your HTML using Grunt.

grunt-bower-requirejs
Automagically wire-up installed Bower components into your RequireJS config.

grunt bower-task
Grunt plugin to automate Bower commands; allow the configuration of the files needed allowing to filter out the minimal in the project.

kooshy-fe
Integrates a web-based interface for Bower

ruby-bower
Ruby binding for Bower commands (Uses node/execjs instead of shelling out)

SpBowerBundle
Bower integration with symfony2.

sublime-bower
A Sublime text-editor plugin for Bower.

Search seems to be down (and I'm not sure how this is backed by GitHub Pages):

$ wget -S http://bower.io/search/
--2014-06-17 11:23:25--  http://bower.io/search/
Resolving bower.io... 192.30.252.153, 192.30.252.154
Connecting to bower.io|192.30.252.153|:80... connected.
HTTP request sent, awaiting response... 
  HTTP/1.1 404 Not Found
  Server: GitHub.com
  Date: Tue, 17 Jun 2014 18:23:55 GMT
  Content-Type: text/html; charset=utf-8
  Content-Length: 9371
2014-06-17 11:23:25 ERROR 404: Not Found.

Google's cache has a working version at Jun 15, 2014 17:14:40 GMT.

The website should have docs on how to use Bower, workflow, etc.

Also basic docs on publishing to Bower; complete details should be within the CLI tool's help pages (separate issue).

Since we don't have a blog, we should have a place to post snippets of news with a RSS feed. Twitter is nice but user needs to parse through non-news content. Can just be a single page that we keep appending to.

Example content:

• CLI releases
• Planned downtime
• Deprecation (e.g. just received email from heroku today that bower.heroku.com routing will be deprecated)

"you're are" -> "you are"

https://github.com/bower/bower.github.io/blame/0c6e2cd4e2491e1a8568c8f25ea1f55e907587e9/docs/creating-packages.md#L91

http://bower.io/docs/config/#registry

The registry config. Can be an object or a string. If a string is used, all the property values _bellow_...

...should be "below"

The webpage is not updated to the latest README in master.

For example, in the website I can read:

All package contents are installed in the components directory by default. You should never directly modify the contents of this directory.

The README.md is like this:

All package contents are installed in the bower_components directory by default. You should never directly modify the contents of this directory as any changes can and probably will get removed.

I think this is a major problem.

Hi,

I published a package on bower months ago, and it still doen't show up in the website's list…
The project is hosted on Bitbucket : https://bitbucket.org/lsystems/angular-extend-promises

The repository works fine, and the install via bower too, so I think it is correctly registered.

I have no clue why it doesn't show up on bower.io though…

Thank you in advance for your help!

I added a promotion for the Bower shirts on the homepage. This is a limited time offer, ending December 8th — at which point I'll remove it.

These shirts are sold at cost, so I'm not trying to make a buck. Just trying to spread the Bower 🐦 .

It looks like this

1. "Defaults to paths that suite the OS/platform."
   Is "suite" a typo?
2. "Valid keys are cache, registry, links, completion."
   And then the example shows a key of "packages."

These are on the page at http://bower.io/docs/config/.

The docs for configuring bower should have examples.

storage

Object

Where to store persistent data, such as cache, needed by bower. Defaults to paths that suite the OS/platform. Valid keys are cache, registry, links, completion.

{
    "storage" : {
         "packages" : "~/.bower/cache",
         "registry" : "~/.bower/registry"
         ... 
    }
}

A bigger and more visible button (instead of just a link) for search page on http://bower.io/ would be nice.

It could also be a form, that submits to /search?q=keyword.

Hi,

To improve the flow of adding Bower to a solution can the dependency install directory be specified within the bower init process. I'd like to see the directory value from .bowerrc being integrated with the bower.json. One config file to rule them all. I may have missed something from the intended purpose of .bowerrc, perhaps in a larger set of nested solutions or the inheriting properties but for working on different projects with different standards it would be easier to have one process to create all the configuration needed.

Thanks,
Mike

(I opened a documentation issue in the bower repo.)

Bower says my package is v1.0.0 in one place, but it was resolved as v2.0.0. When I check the bower.json and package.json, both files have 2.0.0...thoughts?

to let bower run in a production environment how are security issues, security advisories, security updates handled, notified, ...? could not find any information to security at all....

The code font on bower's homepage has the following CSS:
pre, code { font-family: Consolas,Menlo,Courier,monospace; }
GNU/Linux users tend not to have Consolas (primarily Windows) or Menlo (primarily OS X) but do have Courier, but Courier is really ugly for code, both in my opinion and the opinion of many people that work with code. I think it would be better without the "Courier" and to leave monospace or replace it with a nicer font that generally comes with default GNU/Linux distributions, like Terminus or something. I was just browsing the site and this really stuck out to me lol.

Thanks!

Personally, I think it would be better (and more intuitive) to keep the source of the website in a repository with the same name as the domain and leave bower.github.(com|io) for, maybe, something organization related (e.g.: https://github.com/h5bp/h5bp.github.com).

Would like to see the team's appetite to make things a little more secure. We can get a free cert from StartSSL, however it only allows 1 subdomain. We may need at least:

• stats.bower.io
• search.bower.io (if we decide to move things to be more consistent, see https://github.com/bower/search/issues/37)
• api.bower.io (to be built, currently lives under stats.bower.io/api/, see https://github.com/bower/stats.bower.io/issues/2)

Also maybe this can be used for the CLI tool as well?

from bower/bower#1352

When I specify an endpoint, I want that installed with its dependencies, and nothing else, specifically I don't want the [bower.json] project dependencies installed.

It seems like anchor links aren't being inserted properly, linking AnchorLink instead of the actual link:

It looks fine in most browsers though, for some reason.

The "Creating packages" docs say that packages should be tagged with semver Git tags. I think this section of the docs should clarify whether the tag name can have a prefix and what prefixes are allowed (e.g. v1.0.0). I don't think semver allows a prefix, but it seems like Bower permits one (see this issue, for example). So I think the docs should clarify this. Thanks.

Re bower/bower#886

@wibblymat and @satazor concurred that PR would be better implemented as a 'Bower plugin', but we have no documentation around what that means, or how to do it.

Some suggested aims of this ticket:

• Tighter branding
• Better introduction to Bower and the actual problem it is solving.
• Expose who is using Bower.
• Expose who is supporting its development.
• Point to a web interface for searching for packages.
• Create distinct sections that can house API documentation overviews (not a replacement for the CLI help pages).
• Example workflows (possibly including use of other tools).
• Examples of well authored packages.
• Other things? Fewer things?

It would be good to have all this done in the open, on GitHub. And to keep the site (and working on it) simple. I'm happy to help out as much as possible. Feel free to jump in with some thoughts.

cc @isaacdurazo

Draft site IA:

• Home
  • What is Bower? Why should I care?
  • Who is using Bower?
• Getting started
  • Installing Bower
  • Installing components
  • Basic use of .bowerrc
  • Finding components
• Creating Bower packages
  • What is a Bower package? (e.g., should be small, focused).
  • Use of bower.json.
  • No-registration required for simple packages - GitHub shorthand.
  • Registering a Git endpoint for your package so it shows up for bower search.
  • Publishing assets directly to the Bower registry (why and how).
• Advanced use
  • Setting up and referencing a private Bower registry (for companies).
  • Use of programmatic API.
• Commands reference
• Workflows
  • How can I use Bower with X?
  • Grunt
  • Require JS / AMD
• Web site press kit area (For various logo downloads)
• Bower badges (benschwarz.github.io/bower-badges)
• FAQ
  • Windows Git gotcha (this could probably go under "Getting started"

It's a great service to the developer community if tool originators present, up front, a concise and clear summary of what their tool is and does, so developers can quickly assess whether it's relevant to their field of work, and whether to invest further time learning it.

I think the Bower home page http://bower.io/ could use some refinement to that end. Here's a blow by blow commentary on me trying to get traction on what it says currently:

Bower A package manager for the web

Nice hook, but takes in pretty broad possibilities. We see that Bower has something to do with package management, and something to do with the web, which is a vast scope. We'll be needing more specifics...

Web sites are made of lots of things — frameworks, libraries, assets, utilities, and rainbows.

This description fits lots of items that run on the server, and on the browser; does Bower deal with all these? Is this saying Bower can "manage" any of Django, MySQL, jquery, or angular (as examples).

Bower manages all these things for you.

Well, it says "all", but I'm skeptical. From other reading, I'm pretty sure Bower actually doesn't manage things that run on the server, but perhaps I'm wrong. But now, what is the scope of "manage"?

Bower works by fetching and installing packages from all over, taking care of hunting, finding, downloading, and saving the stuff you’re looking for.

Again, Bower claims to install, but "hunting, finding, downloading, saving" doesn't include "configuring". So still not sure what "install" means here.

From other reading, I get the impression that Bower doesn't have features to install anything to run locally. Instead, it marshals files ready for the server to deliver them to clients, which is where they actually run, or appear, etc. Maybe in some sphere of web development, that's called "installing", but readers with broader experience might not know that this is the limited definition intended.

And what is the scope of "all over"?

Bower keeps track of these packages in a manifest file, bower.json. So.... that's like a repo database or something? And there's just one bower.json for all packages?

How you use packages is up to you. This is meaningless as exposition of Bower. Clearly it's only "up to me" to choose among things that Bower facilitates. Which is what I'm here to find out, and was hoping that a sentence right about now would tell me.

Bower provides hooks to facilitate using packages in your tools and workflows. The hooks are for using the packages? Or is this trying to say that Bower has an API for using Bower to perform package-related tasks?

Bower is optimized for the front-end. Does this mean that sometimes Bower can be used for "installing" back-end things, but mostly it's for "installing" front-end components? Again, I'm pretty sure that Bower is for preparing delivery only of front-end components. So the "optimized" claim doesn't seem to add to the discussion, and only makes a muddled suggestion that maybe there are other less-optimized uses.

Bower uses a flat dependency tree, requiring only one version for each package, Does "flat dependency tree" bear somehow on "one version per package"? Or is this a non-sequitur? What does "flat dependency tree" even mean?

reducing page load to a minimum. Possibly "page load time"?

Install Bower
Bower is a command line utility. Install it with npm. OK

**npm install -g bower

Bower requires Node and npm and Git.** These two lines are in reverse order that they are needed.

Getting started
Install packages
Install packages with bower install. Bower installs packages to bower_components/.

So, if components are "installed" to bower_components/ that suggests that "install" does not mean install for local use, right? So we infer that Bower's purpose is to gather files for delivery to the browser, and not at all for local use. But readers shouldn't have to reverse engineer this understanding, if indeed it's even true.

bower install A package can be a GitHub shorthand, a Git endpoint, a URL, and more. Read more about bower install

This is likely supposed to say that the argument can be any of those things, not that packages themselves can be these things.

Search packages
Search Bower packages and find the registered package names for your favorite projects.

This seems confusing. Earlier the narrative says Bower can get stuff from "all over". Yet here it introduces "Bower packages". Does Bower require that candidate packages be pre-cooked in some Bower-specific way? I'm pretty sure not, but then what does this mean?

Save packages Save your packages to bower.json with bower init.

Pretty sure that the packages are not saved to bower.json. I suspect only references to the package are saved to bower.json, while the packages themselves remain in bower_components. But, that's a guess.

Use packages
How you use packages is up to you.

Again, this is meaningless as exposition of Bower, and simply frustrates anyone who doesn't already know what Bower does.

Use Bower together with Grunt, RequireJS, Yeoman, and lots of other tools or build your own workflow with the API.

It's useful to know the ecosystem that Bower might work with, and it would be even better if the sentence let us know whether Bower would play a subsidiary role or a supervisory role to these other tools.

There you have it. There's a lot of ambiguity which I think could be ironed out to very good effect.

We should have an official recommendation / policy for these. Let me know if this is ok and I'll submit a PR.

• Namespacing - we don't support github style namespacing, however people are encouraged to namespace related modules with -. E.g. angular- or Polymer's paper-
• Anti-squatting - I recommend we adopt npm's policy - publish code or move out of the way
• Name transfer / IP & other disputes - try to work it out with the owner first, if no resolution, submit a ticket in the Bower Registry repo and the core team will sort it out

This should go into the Creating Package page.

Provide a directory structure, volunteers will participate in.

Thanks.

请考虑支持多语言环境

只要提供目录结构即可，有自愿者会参与的。

感谢。

제 목록 구조 하 면 되 는 자원 봉사자 들이 참여 할 한 다.

감사 드 린 다.