# For Comment: Gem Browser

**URL:** <https://rubytalk.org/t/for-comment-gem-browser/45232>\
**Category:** ruby-talk\
**Created:** [17 March 2008 03:20 UTC](https://rubytalk.org/t/for-comment-gem-browser/45232 "2008-03-17T03:20:54Z")\
**Posts on this page:** 20\
**Page:** 1

<div class="post-metadata">

**Author:** ![Steve\_Ross](https://avatars.discourse-cdn.com/v4/letter/s/3d9bf3/32.png) [@Steve\_Ross](https://rubytalk.org/u/Steve_Ross)\
**Post date:** [17 March 2008 03:20 UTC](https://rubytalk.org/t/for-comment-gem-browser/45232/1 "2008-03-17T03:20:54Z")

</div>

It occurred to me that the way I was finding Gems was by making notes of what I saw coming across this list and various other disorganized ways. That didn't seem too bright and searching RubyForge didn't always turn things up as I expected. So, having a dull hour or two, I cobble a browser together.

Obviously, this is a really crude attempt and I'm just looking for comments regarding whether it's useful:

[http://gemmy.calicowebdev.com](http://gemmy.calicowebdev.com)

It's only running on one mongrel and I've done little to optimize it. The searching is performed using ferret. Comments welcome on this as well.

Currently searchable fields are gem name, author, and description. If this seems like a bright idea, I'll probably add user annotations and tags for starters and make them searchable as well.

Thanks in advance for your comments.

---

<div class="post-metadata">

**Author:** ![Chad\_Perrin1](https://avatars.discourse-cdn.com/v4/letter/c/977dab/32.png) [@Chad\_Perrin1](https://rubytalk.org/u/Chad_Perrin1)\
**Post date:** [18 March 2008 05:53 UTC](https://rubytalk.org/t/for-comment-gem-browser/45232/2 "2008-03-18T05:53:54Z")

</div>

It appears to be broken. There's no submit button, and hitting Enter  
doesn't do anything when I've entered a term in the Find box. Am I  
misusing it somehow?

> **···**
>
> On Mon, Mar 17, 2008 at 12:20:54PM +0900, s.ross wrote:
> 
> > It occurred to me that the way I was finding Gems was by making notes  
> > of what I saw coming across this list and various other disorganized  
> > ways. That didn't seem too bright and searching RubyForge didn't  
> > always turn things up as I expected. So, having a dull hour or two, I  
> > cobble a browser together.
> > 
> > Obviously, this is a really crude attempt and I'm just looking for  
> > comments regarding whether it's useful:
> > 
> > [http://gemmy.calicowebdev.com](http://gemmy.calicowebdev.com)
> > 
> > It's only running on one mongrel and I've done little to optimize it.  
> > The searching is performed using ferret. Comments welcome on this as  
> > well.
> > 
> > Currently searchable fields are gem name, author, and description. If  
> > this seems like a bright idea, I'll probably add user annotations and  
> > tags for starters and make them searchable as well.
> > 
> > Thanks in advance for your comments.
> 
> --  
> CCD CopyWrite Chad Perrin [[http://ccd.apotheon.org](http://ccd.apotheon.org) ]  
> Isaac Asimov: "Part of the inhumanity of the computer is that, once it is  
> completely programmed and working smoothly, it is completely honest."

---

<div class="post-metadata">

**Author:** ![Martin\_DeMello](https://avatars.discourse-cdn.com/v4/letter/m/f17d59/32.png) [@Martin\_DeMello](https://rubytalk.org/u/Martin_DeMello)\
**Post date:** [18 March 2008 17:49 UTC](https://rubytalk.org/t/for-comment-gem-browser/45232/3 "2008-03-18T17:49:26Z")

</div>

I don't understand the results it gives me for 'gtk'

martin

> **···**
>
> On Sun, Mar 16, 2008 at 8:20 PM, s.ross \<cwdinfo@gmail.com\> wrote:
> 
> > [http://gemmy.calicowebdev.com](http://gemmy.calicowebdev.com)
> > 
> > It's only running on one mongrel and I've done little to optimize it.  
> > The searching is performed using ferret. Comments welcome on this as  
> > well.
> > 
> > Currently searchable fields are gem name, author, and description. If  
> > this seems like a bright idea, I'll probably add user annotations and  
> > tags for starters and make them searchable as well.

---

<div class="post-metadata">

**Author:** ![Paul\_Brannan](https://avatars.discourse-cdn.com/v4/letter/p/76d3ee/32.png) [@Paul\_Brannan](https://rubytalk.org/u/Paul_Brannan)\
**Post date:** [24 March 2008 15:07 UTC](https://rubytalk.org/t/for-comment-gem-browser/45232/4 "2008-03-24T15:07:36Z")

</div>

Looks interesting to me, but it's really more a search tool than a  
browser.

I think it would be nifty to use the rubyforge trove categorization to  
create a tree view of all the gems.

Paul

> **···**
>
> On Mon, Mar 17, 2008 at 12:20:54PM +0900, s.ross wrote:
> 
> > Obviously, this is a really crude attempt and I'm just looking for  
> > comments regarding whether it's useful:
> > 
> > [http://gemmy.calicowebdev.com](http://gemmy.calicowebdev.com)

---

<div class="post-metadata">

**Author:** ![Thomas\_P](https://avatars.discourse-cdn.com/v4/letter/t/8797f3/32.png) [@Thomas\_P](https://rubytalk.org/u/Thomas_P)\
**Post date:** [24 March 2008 18:43 UTC](https://rubytalk.org/t/for-comment-gem-browser/45232/5 "2008-03-24T18:43:07Z")

</div>

it does not find " Summary: roo can access the contents of OpenOffice-,  
Excel- or Google-Spreadsheets" if i search for 'spreadsheet'.

-Thomas

> **···**
>
> On 17/03/2008, s.ross \<cwdinfo@gmail.com\> wrote:
> 
> > Obviously, this is a really crude attempt and I'm just looking for  
> > comments regarding whether it's useful:
> > 
> > [http://gemmy.calicowebdev.com](http://gemmy.calicowebdev.com)
> 
> --  
> Thomas Preymesser  
> thopre@gmail.com  
> Telefon: 030 - 49 78 37 06
> 
> > **[thopre](https://thopre.wordpress.com/)**
> >
> > Entwicklungen für die Bühne des Lebens

---

<div class="post-metadata">

**Author:** ![Chad\_Perrin1](https://avatars.discourse-cdn.com/v4/letter/c/977dab/32.png) [@Chad\_Perrin1](https://rubytalk.org/u/Chad_Perrin1)\
**Post date:** [25 March 2008 06:17 UTC](https://rubytalk.org/t/for-comment-gem-browser/45232/6 "2008-03-25T06:17:58Z")

</div>

I posted something about this in my personal weblog, and got this  
response:

&nbsp;&nbsp;[http://sob.apotheon.org/?p=375#comment-330556](http://sob.apotheon.org/?p=375#comment-330556)

The text of it says:

&nbsp;&nbsp;I also use the full file navigation functionality, which lets me  
&nbsp;&nbsp;navigate the distribution's directory structure and look inside the  
&nbsp;&nbsp;various files. Especially handy for reading full source code. I'd love  
&nbsp;&nbsp;to see that included here.

I just figured I'd pass it along.

> **···**
>
> On Mon, Mar 17, 2008 at 12:20:54PM +0900, s.ross wrote:
> 
> > Thanks in advance for your comments.
> 
> --  
> CCD CopyWrite Chad Perrin [[http://ccd.apotheon.org](http://ccd.apotheon.org) ]  
> print substr("Just another Perl hacker", 0, -2);

---

<div class="post-metadata">

**Author:** ![Jan\_Svitok](https://avatars.discourse-cdn.com/v4/letter/j/cab0a1/32.png) [@Jan\_Svitok](https://rubytalk.org/u/Jan_Svitok)\
**Post date:** [25 March 2008 08:01 UTC](https://rubytalk.org/t/for-comment-gem-browser/45232/7 "2008-03-25T08:01:38Z")

</div>

Another thing that would be nice: hyperlinked dependencies

> **···**
>
> On Mon, Mar 17, 2008 at 4:20 AM, s.ross \<cwdinfo@gmail.com\> wrote:
> 
> > Thanks in advance for your comments.

---

<div class="post-metadata">

**Author:** ![Chad\_Perrin1](https://avatars.discourse-cdn.com/v4/letter/c/977dab/32.png) [@Chad\_Perrin1](https://rubytalk.org/u/Chad_Perrin1)\
**Post date:** [18 March 2008 06:11 UTC](https://rubytalk.org/t/for-comment-gem-browser/45232/8 "2008-03-18T06:11:32Z")

</div>

Okay, so I figured out it doesn't work without accepting a cookie (a  
misfeature, in my opinion). There's also the problem that there's no way  
to link directly to a given search, as far as I can tell.

Is there some reason that a blank Find field shows 25 results?

I like having an online gem search tool with a clean interface like this.

> **···**
>
> On Tue, Mar 18, 2008 at 02:53:54PM +0900, Chad Perrin wrote:
> 
> > On Mon, Mar 17, 2008 at 12:20:54PM +0900, s.ross wrote:  
> > \> It occurred to me that the way I was finding Gems was by making notes  
> > \> of what I saw coming across this list and various other disorganized  
> > \> ways. That didn't seem too bright and searching RubyForge didn't  
> > \> always turn things up as I expected. So, having a dull hour or two, I  
> > \> cobble a browser together.  
> > \>  
> > \> Obviously, this is a really crude attempt and I'm just looking for  
> > \> comments regarding whether it's useful:  
> > \>  
> > \> [http://gemmy.calicowebdev.com](http://gemmy.calicowebdev.com)  
> > \>  
> > \> It's only running on one mongrel and I've done little to optimize it.  
> > \> The searching is performed using ferret. Comments welcome on this as  
> > \> well.  
> > \>  
> > \> Currently searchable fields are gem name, author, and description. If  
> > \> this seems like a bright idea, I'll probably add user annotations and  
> > \> tags for starters and make them searchable as well.  
> > \>  
> > \> Thanks in advance for your comments.
> > 
> > It appears to be broken. There's no submit button, and hitting Enter  
> > doesn't do anything when I've entered a term in the Find box. Am I  
> > misusing it somehow?
> 
> --  
> CCD CopyWrite Chad Perrin [[http://ccd.apotheon.org](http://ccd.apotheon.org) ]  
> awj @reddit: "The terms never and always are never always true."

---

<div class="post-metadata">

**Author:** ![Steve\_Ross](https://avatars.discourse-cdn.com/v4/letter/s/3d9bf3/32.png) [@Steve\_Ross](https://rubytalk.org/u/Steve_Ross)\
**Post date:** [18 March 2008 18:04 UTC](https://rubytalk.org/t/for-comment-gem-browser/45232/9 "2008-03-18T18:04:55Z")

</div>

Gtk is a bit of a special case. Anyone who understands ferret, please chime in here. My understanding is that shorter words tend to get less weight. I get one hit for "gtk" and two for "\*gtk\*". I'd like to improve this kind of case. Part of this is how the gems are named and part that the string is short. Again, if anyone knows ferret and has suggestions, I'm all ears.

> **···**
>
> On Mar 18, 2008, at 10:49 AM, Martin DeMello wrote:
> 
> > On Sun, Mar 16, 2008 at 8:20 PM, s.ross \<cwdinfo@gmail.com\> wrote:
> > 
> > > [http://gemmy.calicowebdev.com](http://gemmy.calicowebdev.com)
> > > 
> > > It's only running on one mongrel and I've done little to optimize it.  
> > > The searching is performed using ferret. Comments welcome on this as  
> > > well.
> > > 
> > > Currently searchable fields are gem name, author, and description. If  
> > > this seems like a bright idea, I'll probably add user annotations and  
> > > tags for starters and make them searchable as well.
> > 
> > I don't understand the results it gives me for 'gtk'
> > 
> > martin

---

<div class="post-metadata">

**Author:** ![Steve\_Ross](https://avatars.discourse-cdn.com/v4/letter/s/3d9bf3/32.png) [@Steve\_Ross](https://rubytalk.org/u/Steve_Ross)\
**Post date:** [24 March 2008 20:59 UTC](https://rubytalk.org/t/for-comment-gem-browser/45232/10 "2008-03-24T20:59:12Z")

</div>

Yes, that's what it is right now. My hope is that one day it can evolve into a bit more. I have some thoughts about getting the rdocs into the mix. What would make it more browser-like to you?

> **···**
>
> On Mar 24, 2008, at 8:07 AM, Paul Brannan wrote:
> 
> > Looks interesting to me, but it's really more a search tool than a  
> > browser.

---

<div class="post-metadata">

**Author:** ![Steve\_Ross](https://avatars.discourse-cdn.com/v4/letter/s/3d9bf3/32.png) [@Steve\_Ross](https://rubytalk.org/u/Steve_Ross)\
**Post date:** [24 March 2008 21:07 UTC](https://rubytalk.org/t/for-comment-gem-browser/45232/11 "2008-03-24T21:07:04Z")

</div>

Yes, this is a limitation of the tokenizer in ferret. If you type in spreadsheet\* you will get the results you expect (I think). I'm considering making wildcard searching the default, but many developers feel that they are "smart enough" to tweak their own searches, which is why I left it with default behavior initially.

roo contains the word Spreadsheets, but it is in the context "Google-Spreadsheets", so it is not tokenized as Google, Spreadsheet but rather as "Google-Spreadsheet". Similarly, for spreadsheet-excel, the tokenizer kills it.

If you are looking for the most results, use a wildcard at either end (\*spreadsheet\*). If you think you're close, but not sure, use fuzzy matching (spreadsheet~), and refine your searches like this: spreadsheet~ AND open\*.

--steve

> **···**
>
> On Mar 24, 2008, at 11:43 AM, Thomas Preymesser wrote:
> 
> > it does not find " Summary: roo can access the contents of OpenOffice-,  
> > Excel- or Google-Spreadsheets" if i search for 'spreadsheet'.

---

<div class="post-metadata">

**Author:** ![Steve\_Ross](https://avatars.discourse-cdn.com/v4/letter/s/3d9bf3/32.png) [@Steve\_Ross](https://rubytalk.org/u/Steve_Ross)\
**Post date:** [25 March 2008 16:25 UTC](https://rubytalk.org/t/for-comment-gem-browser/45232/12 "2008-03-25T16:25:23Z")

</div>

I've put it in the TODO list. It's another set of tables and another fairly extensive scan of the database to construct this, so it may take a while to find its way in, but I agree about its value.

> **···**
>
> On Mar 25, 2008, at 1:01 AM, Jano Svitok wrote:
> 
> > On Mon, Mar 17, 2008 at 4:20 AM, s.ross \<cwdinfo@gmail.com\> wrote:
> > 
> > > Thanks in advance for your comments.
> > 
> > Another thing that would be nice: hyperlinked dependencies

---

<div class="post-metadata">

**Author:** ![Steve\_Ross](https://avatars.discourse-cdn.com/v4/letter/s/3d9bf3/32.png) [@Steve\_Ross](https://rubytalk.org/u/Steve_Ross)\
**Post date:** [18 March 2008 07:59 UTC](https://rubytalk.org/t/for-comment-gem-browser/45232/13 "2008-03-18T07:59:00Z")

</div>

> Okay, so I figured out it doesn't work without accepting a cookie (a  
> misfeature, in my opinion). There's also the problem that there's no way  
> to link directly to a given search, as far as I can tell.

Yeah, there's really no reason it needs to set a cookie right now. But... I'm thinking there may be a need in the future for session storage and that will require cookies.

Regarding linking to a given search, I'm pretty sure I don't understand... do you mean like a permalink to a given result set?

> Is there some reason that a blank Find field shows 25 results?

Programmer decision 🙂 I decided to make a blank field equal to \* and only show 25 results at a time. Again, I had to take a stab at it, and that was my initial take.

> I like having an online gem search tool with a clean interface like this.

Thanks. I was totally blown away by what I learned by spielunking the gems just a little with this search. I hope it's as much fun for others as it is for me.

Please let me know if you have other thoughts on this.

> **···**
>
> On Mar 17, 2008, at 11:11 PM, Chad Perrin wrote:

---

<div class="post-metadata">

**Author:** ![Jan\_Svitok](https://avatars.discourse-cdn.com/v4/letter/j/cab0a1/32.png) [@Jan\_Svitok](https://rubytalk.org/u/Jan_Svitok)\
**Post date:** [25 March 2008 16:32 UTC](https://rubytalk.org/t/for-comment-gem-browser/45232/14 "2008-03-25T16:32:33Z")

</div>

Couldn't you just issue a query containing the dependency name?

> **···**
>
> On Tue, Mar 25, 2008 at 5:25 PM, s.ross \<cwdinfo@gmail.com\> wrote:
> 
> > On Mar 25, 2008, at 1:01 AM, Jano Svitok wrote:
> > 
> > \> On Mon, Mar 17, 2008 at 4:20 AM, s.ross \<cwdinfo@gmail.com\> wrote:  
> > \>\> Thanks in advance for your comments.  
> > \>  
> > \> Another thing that would be nice: hyperlinked dependencies
> > 
> > I've put it in the TODO list. It's another set of tables and another  
> > fairly extensive scan of the database to construct this, so it may  
> > take a while to find its way in, but I agree about its value.

---

<div class="post-metadata">

**Author:** ![Chad\_Perrin1](https://avatars.discourse-cdn.com/v4/letter/c/977dab/32.png) [@Chad\_Perrin1](https://rubytalk.org/u/Chad_Perrin1)\
**Post date:** [19 March 2008 00:45 UTC](https://rubytalk.org/t/for-comment-gem-browser/45232/15 "2008-03-19T00:45:14Z")

</div>

> \>Okay, so I figured out it doesn't work without accepting a cookie (a  
> \>misfeature, in my opinion). There's also the problem that there's  
> \>no way  
> \>to link directly to a given search, as far as I can tell.
> 
> Yeah, there's really no reason it needs to set a cookie right now.  
> But... I'm thinking there may be a need in the future for session  
> storage and that will require cookies.

Then refusing cookies should break session storage, but not basic search  
functionality.

> Regarding linking to a given search, I'm pretty sure I don't  
> understand... do you mean like a permalink to a given result set?

Yes.

> \>Is there some reason that a blank Find field shows 25 results?
> 
> Programmer decision 🙂 I decided to make a blank field equal to \* and  
> only show 25 results at a time. Again, I had to take a stab at it, and  
> that was my initial take.

You might want to indicate what's going on. I think that's a pretty neat  
idea for how to handle it, but I might suggest something like making the  
selected results (pseudo-)random and starting them off with a heading  
that says "25 Random Gems:".

> \>I like having an online gem search tool with a clean interface like  
> \>this.
> 
> Thanks. I was totally blown away by what I learned by spielunking the  
> gems just a little with this search. I hope it's as much fun for  
> others as it is for me.
> 
> Please let me know if you have other thoughts on this.

Something I didn't say explicitly: I'm sure you'll be adding some more to  
the interface if you keep working on this. Please don't clutter things  
up too much, though. As I said, I like the clean interface, and making  
it look like the search results at RubyForge (for instance) would be a  
Bad Thing. Take your cues from the main page of Google, and not Yahoo!,  
or from reddit, and not Digg.

> **···**
>
> On Tue, Mar 18, 2008 at 04:59:00PM +0900, s.ross wrote:
> 
> > On Mar 17, 2008, at 11:11 PM, Chad Perrin wrote:
> 
> --  
> CCD CopyWrite Chad Perrin [[http://ccd.apotheon.org](http://ccd.apotheon.org) ]  
> awj @reddit: "The terms never and always are never always true."

---

<div class="post-metadata">

**Author:** ![Steve\_Ross](https://avatars.discourse-cdn.com/v4/letter/s/3d9bf3/32.png) [@Steve\_Ross](https://rubytalk.org/u/Steve_Ross)\
**Post date:** [25 March 2008 20:14 UTC](https://rubytalk.org/t/for-comment-gem-browser/45232/16 "2008-03-25T20:14:18Z")

</div>

At first blush, yes. But if it's implemented as a many-to-many relation, I can develop richer information like what depends on a particular gem. A direct query might be a good initial implementation though.

> **···**
>
> On Mar 25, 2008, at 9:32 AM, Jano Svitok wrote:
> 
> > On Tue, Mar 25, 2008 at 5:25 PM, s.ross \<cwdinfo@gmail.com\> wrote:
> > 
> > > On Mar 25, 2008, at 1:01 AM, Jano Svitok wrote:
> > > 
> > > > On Mon, Mar 17, 2008 at 4:20 AM, s.ross \<cwdinfo@gmail.com\> wrote:
> > > > 
> > > > > Thanks in advance for your comments.
> > > > 
> > > > Another thing that would be nice: hyperlinked dependencies
> > > 
> > > I've put it in the TODO list. It's another set of tables and another  
> > > fairly extensive scan of the database to construct this, so it may  
> > > take a while to find its way in, but I agree about its value.
> > 
> > Couldn't you just issue a query containing the dependency name?

---

<div class="post-metadata">

**Author:** ![Steve\_Ross](https://avatars.discourse-cdn.com/v4/letter/s/3d9bf3/32.png) [@Steve\_Ross](https://rubytalk.org/u/Steve_Ross)\
**Post date:** [19 March 2008 07:06 UTC](https://rubytalk.org/t/for-comment-gem-browser/45232/17 "2008-03-19T07:06:01Z")

</div>

Hi--

A few updates to [http://gemmy.calicowebdev.com](http://gemmy.calicowebdev.com).

+ Added description field if one is present. This is typically either blank or the same as the summary, but where it differs, it is expanded on demand.  
+ Added more weight to gem name in an attempt to tweak full-text matching accuracy.  
+ Removed "it just breaks" dependency on cookies.  
+ Added a bit more commentary to the results count to explain why only 25 are shown when 2300 are found.  
+ Added obligatory dumb logo.  
+ Added some "help" text (as if people who need Ruby Gems need help!).

The goal of this tool is to be pretty quick and pretty functional and pretty bare bones. Just the facts.

As always, I'm interested in your comments and suggestions -- particularly whether search is finding what you expect and whether help is helpful or a waste of time.

Thanks

---

<div class="post-metadata">

**Author:** ![Jan\_Svitok](https://avatars.discourse-cdn.com/v4/letter/j/cab0a1/32.png) [@Jan\_Svitok](https://rubytalk.org/u/Jan_Svitok)\
**Post date:** [19 March 2008 09:39 UTC](https://rubytalk.org/t/for-comment-gem-browser/45232/18 "2008-03-19T09:39:23Z")

</div>

You could #uniq the version numbers and list platforms as well.

> **···**
>
> On Wed, Mar 19, 2008 at 8:06 AM, s.ross \<cwdinfo@gmail.com\> wrote:
> 
> > Hi--
> > 
> > A few updates to [http://gemmy.calicowebdev.com](http://gemmy.calicowebdev.com).
> > 
> > + Added description field if one is present. This is typically either  
> > blank or the same as the summary, but where it differs, it is expanded  
> > on demand.  
> > + Added more weight to gem name in an attempt to tweak full-text  
> > matching accuracy.  
> > + Removed "it just breaks" dependency on cookies.  
> > + Added a bit more commentary to the results count to explain why only  
> > 25 are shown when 2300 are found.  
> > + Added obligatory dumb logo.  
> > + Added some "help" text (as if people who need Ruby Gems need help!).
> > 
> > The goal of this tool is to be pretty quick and pretty functional and  
> > pretty bare bones. Just the facts.
> > 
> > As always, I'm interested in your comments and suggestions --  
> > particularly whether search is finding what you expect and whether  
> > help is helpful or a waste of time.

---

<div class="post-metadata">

**Author:** ![Chad\_Perrin1](https://avatars.discourse-cdn.com/v4/letter/c/977dab/32.png) [@Chad\_Perrin1](https://rubytalk.org/u/Chad_Perrin1)\
**Post date:** [19 March 2008 19:53 UTC](https://rubytalk.org/t/for-comment-gem-browser/45232/19 "2008-03-19T19:53:02Z")

</div>

It looks excellent.

+ Description: The way the description has been added is great. The font  
is a touch small, and a way to unexpand it again after expanding might be  
nice, though I suppose the unexpand isn't really important.  
+ Cookies: Glad to hear it doesn't break without cookies now. Thank you.  
Even though I'm accepting cookies from the site, websites that fail  
without cookies bother me -- and I'm sure I'm not the only one.  
+ 25 results: Are you planning to add some functionality for seeing the  
rest of the results at some point, or just expect that people will refine  
their searches somehow if they don't get the results they wanted?  
+ Logo: I like "{ zippy graphic omitted }". Don't take my preference in  
this for objective advice, though.  
+ Help: In my opinion, the way you have the help text presented is  
excellent. Keep it.

I tend to keep JavaScript turned on in my browser, but there are (of  
course) others who do not. The fact the Gemmy Gem Browser is unusable  
without JavaScript seems like an unnecessary limitation. This probably  
isn't the most important problem to fix, but you might consider it for  
the future, in case someone wants to use it from Lynx, or even write a  
command line script in Ruby that accesses results from GGB.

I'll have to give it a more thorough going-over later to see how I like  
the result sets.

Thanks for putting in the time on this.

> **···**
>
> On Wed, Mar 19, 2008 at 04:06:01PM +0900, s.ross wrote:
> 
> > Hi--
> > 
> > A few updates to [http://gemmy.calicowebdev.com](http://gemmy.calicowebdev.com).
> > 
> > + Added description field if one is present. This is typically either  
> > blank or the same as the summary, but where it differs, it is expanded  
> > on demand.  
> > + Added more weight to gem name in an attempt to tweak full-text  
> > matching accuracy.  
> > + Removed "it just breaks" dependency on cookies.  
> > + Added a bit more commentary to the results count to explain why only  
> > 25 are shown when 2300 are found.  
> > + Added obligatory dumb logo.  
> > + Added some "help" text (as if people who need Ruby Gems need help!).
> > 
> > The goal of this tool is to be pretty quick and pretty functional and  
> > pretty bare bones. Just the facts.
> > 
> > As always, I'm interested in your comments and suggestions --  
> > particularly whether search is finding what you expect and whether  
> > help is helpful or a waste of time.
> 
> --  
> CCD CopyWrite Chad Perrin [[http://ccd.apotheon.org](http://ccd.apotheon.org) ]  
> Patrick J. LoPresti: "Emacs has been replaced by a shell script which 1)  
> Generates a syslog message at level LOG\_EMERG; 2) reduces the user's disk  
> quota by 100K; and 3) RUNS ED!!!!!!"

---

<div class="post-metadata">

**Author:** ![Steve\_Ross](https://avatars.discourse-cdn.com/v4/letter/s/3d9bf3/32.png) [@Steve\_Ross](https://rubytalk.org/u/Steve_Ross)\
**Post date:** [19 March 2008 20:05 UTC](https://rubytalk.org/t/for-comment-gem-browser/45232/20 "2008-03-19T20:05:21Z")

</div>

> It looks excellent.
> 
> + Description: The way the description has been added is great. The font  
> is a touch small, and a way to unexpand it again after expanding might be  
> nice, though I suppose the unexpand isn't really important.

\>\> In the queue.

> + Cookies: Glad to hear it doesn't break without cookies now. Thank you.  
> Even though I'm accepting cookies from the site, websites that fail  
> without cookies bother me -- and I'm sure I'm not the only one.

\>\> Point taken.

> + 25 results: Are you planning to add some functionality for seeing the  
> rest of the results at some point, or just expect that people will refine  
> their searches somehow if they don't get the results they wanted?

\>\> Lower in the queue. I personally feel that the more people think  
\>\> about their question, the better their answers will be. I'll consider

> **···**
>
> On Mar 19, 2008, at 12:53 PM, Chad Perrin wrote:  
> \>\> pagination if people think it's important
> 
> > + Logo: I like "{ zippy graphic omitted }". Don't take my preference in  
> > this for objective advice, though.
> 
> \>\> I had to put a dig in 🙂
> 
> > + Help: In my opinion, the way you have the help text presented is  
> > excellent. Keep it.
> 
> \>\> Thanks
> 
> > I tend to keep JavaScript turned on in my browser, but there are (of  
> > course) others who do not. The fact the Gemmy Gem Browser is unusable  
> > without JavaScript seems like an unnecessary limitation. This probably  
> > isn't the most important problem to fix, but you might consider it for  
> > the future, in case someone wants to use it from Lynx, or even write a  
> > command line script in Ruby that accesses results from GGB.
> 
> \>\> This is exposed as a REST service, so anyone can do searches via  
> \>\> a simple HTTP protocol. Pretty cool, eh?

[Next page](https://rubytalk.org/t/for-comment-gem-browser/45232.md?page=2)
