I would like to talk about an issue I've been thinking about for some time. Which is HOW K-Meleon handles search engines. The old method had one disadvantage. You had to manually update your list of search engines with new entries. But it also enjoyed one nice advantage. By right clicking, you were presented with a list of ALL your search engines (in my case that is now 70 engines) to choose from.

The new method is what I would call a "Ring" of 10 entries. You can change the list of ten that are used, but this is now the official method for using search engines with K-Meleon. But you are stuck with only that list unless you change the defaults in configuration.

I have noticed one possible side effect of this new method which is that fewer people seem interested in adding new search engines to the list in Wiki or post info about new search engines in the forum. Is there a cause and effect here? Can't say.

I would like to propose that future distributions of K-Meleon include BOTH the old and the new search engine methodology (which is what I've done with my version of KM 1.0.2).

I do NOT recall that there was a discussion about the changeover to the "ring" method, merely that it was something that was developed, looked good and was incorporated without much or any discussion.


What do you think?

N

Quote

I do NOT recall that there was a discussion about the changeover to the "ring" method, merely that it was something that was developed, looked good and was incorporated without much or any discussion.

I do not recall that this is a democracy. I don't like to make many words now, I just like to please you to examine the next release of km 1.1 and to see what the "ring" method evolved to. And then tell me that you had foreseen that in a discussion one year ago!

I use scroogle.org.

@wildbill-*.org=ok & kko "could" mean how easy we will add engines with next... ;-)

p.s. I'd prefer meritocracy any day.

kko,

You do not recall that this is a democracy? Not exactly, but not exactly a dictatorship either. I've been around here for a little while and my memory tells me that we are a community where discussion about proposed features has been one of the reasons why we have enjoyed a great reputation over the years.

N



Edited 2 time(s). Last edit at 03/11/2007 01:33AM by ndebord.

wildbill,

I used to use scroogle as my default search engine, but when I found Kart00, I changed over. As I said before, I have 70 search engines via the right click method and use different ones as needed.

N

kko,

Don't get me wrong here. I like many of the macro changes that have been developed, it's just that I would like to see a little more discussion about things that show up in the browser BEFORE they show up!

<wry grin>

N

kko,

How difficult would it be for someone to change the number of displayed entries?

Quote
ndebord
a little more discussion

IMHO it is a big improvement that we now have devs that come here and talk to us.
We should not forget that - and remember what it was like - when we had none at all.

Time is a scarce commodity. I would rather not have long discussions.
And for me it is ok - when our devs implement the things they want and instead of the things I would like. I just change it to my liking as soon as the browser is passed out / well, inside the limits set by my lack of knowledge

BTW. I think that we have got now many of the things that we had always wanted - because our devs also want a better browser.



Edited 1 time(s). Last edit at 03/11/2007 01:29PM by guenter.

Guenter,

I agree with you to this degree. When there is a major change in the browser and no discussion before hand, you lack the all important feedback to let you (as a developer) know whether or not you have actually improved things or made things worse. Feedback can tell you if you have made a misstep before the code goes Gold (as we used to say).

And I too am much happier now about how many people are involved in developing K-Meleon. I remember all too well when we had to petition Ulf to respond and relied heavily upon the good graces of Andrew to act as the go-between between the K-Meleon community and the developer.

N

rmn,

rmn>> How difficult would it be for someone to change the number of displayed entries?

Yes, that thought occurred to me also. A very good point.

N

Quote
rmn
kko,

How difficult would it be for someone to change the number of displayed entries?

:/ Never tried one of the recent builds?

Quote
ndebord
it's just that I would like to see a little more discussion about things that show up in the browser BEFORE they show up!

I don't understand how you want to discuss something BEFORE anybody has seen it. Especially, when you don't have any clue what's technically possible and what's not. Sounds like having blindmen discussing about colors...

Quote
ndebord
When there is a major change in the browser and no discussion before hand, you lack the all important feedback to let you (as a developer) know whether or not you have actually improved things or made things worse. Feedback can tell you if you have made a misstep before the code goes Gold (as we used to say).

What you say now doesn't fit together with what you've said above. How should a developer get feedback on something that he hasn't implemented yet and that he hasn't shown to the public yet? That's why I implement my ideas first. I prefer to have people discuss things they can see and try in practice. When I publish something new, I understand it as a proposal, nothing is set in stone until the final release. I've said that earlier. I listen to critique and to improvement requests. But when there's not more discussion among and feedback from beta testers than this, you can hardly blame me for that. I've taken your critique serious and I've modified my code in order to allow you (and whoever may need it) to go on using the 0.9 search macros.

Quote
ndebord
You do not recall that this is a democracy? Not exactly, but not exactly a dictatorship either.

I'm indeed not willing to cancel a feature I've already spent a lot of time on, just because one single user doesn't like it. If you want to call me a dictator for that, I really don't care.

kko,

If you don't get the gist of my point, there is no point in discussing it further.

N

kko,

FWIW, nobody, including myself, is asking you to remove a feature you spent time on developing. A feature that I think is very good and improves the browser and I do appreciate that you modified the code to allow the right click function to call up search engines as it was done in KM 0.9 and before. It comes down to having a choice on how to search. My personal preference is the way it was done, where you could call up as many search engines as you cared to load up in the macro file. I now use both methods (your ring) and the older right click and like having the ability to choose between the methods.

It's all about choice.

N

Sorry N., but the more you say the less I understand what you actually want. I'm getting the impression that you're turning yourself like a flag in the wind. This day you say that and another day you say something else.

First, you wanted more discussion about proposed features. But when I joined the discussion and told my opinion, I was all wrong and the discussion was over.

This day, you say it's all about choice. About what choice? In km 1.0x you have the choice to use your old macros. And I've told you weeks (months?) ago that you will have the same choice in km 1.1.

So, what do you actually want, N.? Please make yourself clear what you want, then tell us and then we can talk about it. I wouldn't say that if I weren't interested in your opinion.

kko,

I'm trying to make nice, but you're making it very hard to do so (so if you want to characterize that as my twisting in the breeze, be my guest).

A little history here. When you developed the new search engine method, it was presented as a done deal. End of story. I spoke up then and said why fix something that ain't broke (the old right-click method that brought up all your search engines), and voila... you managed to revive the older system. As for your comment that end users will have the same choice in KM 1.1, I managed to miss that statement from you, but it is indeed a welcome one.

The tone of this conversation started going downhill when you said: "I do not recall that this is a democracy," and I recall saying that if it was not a democracy, neither was it a dictatorship. What it has been, is a community of interested end users, part-time code writers and one developer at a time. A small, but motivated community.

N

I can only guess, but have You all looked at how the finished product works at the current beta? I really think that this is close to ideal and represents what is technically possible for such a small dev crew to create in such a short time.

Maybe we users can take it over for an extended "look" and those that are interested in search engines can add a few few enignes that are not yet included. That is more productive than discussion and finds bugs and usability improvements.

Maybe there is some room for future alteration - but that would require specific comments like add an extra "button for removing engines". That is my only idea because i did not immidiately realize that I can add via buttom AND remove via menue item. And that i can change the number of engines :-). It do not know whether there is a limit - since I did not find it.

Thanks to the new code by kko and the new plugin by Dorian the search does more than I will possibly ever need.

Well, N., since you don't say anything essentially new, I also have to repeat myself more or less:

Quote
ndebord
When you developed the new search engine method, it was presented as a done deal. End of story.

I see that different, as I've already said above. When I presented the new implementation, I actually launched a test balloon. I wasn't sure how the community would take it up since the new implementation had some lacks in comparison to the old one. The few feedback I got, was mainly positive. At least, it was not negative enough to make me think that I made things worse. Due to your's and jsnj's comments, I was able to make the new and the old implementation coexist, and I decided to let this code go Gold. I wanted to get the new implementation tested by more users in order to get more feedback. But I didn't get much more feedback and none of it was negative. So, I have to conclude that the new implementation is accepted. But that's not yet the end of the story.

Quote
ndebord
...that ain't broke

Well, yes, the search engine method that was introduced in km 0.9 wasn't broken, but on the other hand, its implementation left no space for further improvements, especially in regard to usability. I mean, once you have set up your engines, it is alright and highly usabel. But the way you have to set up your engines isn't usable at all, not in my opinion. Users are not used to edit source code in order to configure something as basic as their favorite search engines. And besides that, who says that an implementation of a feature has to persist for all time and can never be changed? Software developement is an evolutionary process. You cannot expect that there's always all perfect right from the beginning. Sometimes, development even leads into a bag end and you have to go back first in order of going forth again. From my point of view, the 0.9 implementation was such a bag end. I stepped back in 1.0 and now, in 1.1, we're going forth again. Starting from Las Vegas (Vorne hui, hinten pfui), we went through the Death Valley and now reached sunny California with the Pacific Ocean opening us the world...

Quote
ndebord
As for your comment that end users will have the same choice in KM 1.1, I managed to miss that statement from you, but it is indeed a welcome one.

I've written a little module were you can put all your search engine macros. It didn't make it into 1.1 beta, but I will post it to the macro library in the wiki soon.

Quote
ndebord
The tone of this conversation started going downhill when you said: "I do not recall that this is a democracy," and I recall saying that if it was not a democracy, neither was it a dictatorship. What it has been, is a community of interested end users, part-time code writers and one developer at a time. A small, but motivated community.

You enroll a nice picture here. I've only one problem with it. Who is the community? From my own experience, our community is an accummulation of individualists with partly extremly diverging interests. When they think that something is wrong, they are all quickly shouting. But when they are asked to do something (e.g. updating the macro library for km 1.1), they all silently flee into their rabbit holes. My name is Rabbit, I am not responsible! That's alright. But, I am a developer, my name is not Rabbit, I am responsible. I take responsibility for what I do and that's why I'm the only one who decides how I do things. And since I'm also a volunteer, I'm also the only one who decides what I do and when I do that. In this sense, I am a dictator and every other developer is that too. Democracy cannot work here, except you want to elect volunteers for dictators.

The community's role and responsibility, in my opinion, is to collect ideas for new features and to give the developers feedback during the months - MONTHS! - BEFORE the new code becomes final. When somebody's coming afterwards saying "Huuu, all changed? Why that?", there's nothing left to say.


So, N., I guess, I've provided you enough ammunition. Now stop to make nice and say what think!

Quote
kko
:/ Never tried one of the recent builds?

Oops, I thought I had to fiddle with about:config, macros, or even kmPrefs.jar. :-P

kko,

<<You enroll a nice picture here. I've only one problem with it. Who is the community? From my own experience, our community is an accummulation of individualists with partly extremly diverging interests. When they think that something is wrong, they are all quickly shouting. But when they are asked to do something (e.g. updating the macro library for km 1.1), they all silently flee into their rabbit holes. My name is Rabbit, I am not responsible! That's alright. But, I am a developer, my name is not Rabbit, I am responsible. I take responsibility for what I do and that's why I'm the only one who decides how I do things. And since I'm also a volunteer, I'm also the only one who decides what I do and when I do that. In this sense, I am a dictator and every other developer is that too. Democracy cannot work here, except you want to elect volunteers for dictators. >>

Hey, everybody to their own philosophy. In my time here, I've seen several coders of all stripes and persuasions with differing perspectives about how to proceed. As for users, some have contributed more than others and so too has the level and tone of the discourse ebbed and flowed. That sense of community I see is perhaps just a personal sentiment, but everybody sees things through their own eyes.

N

kko,


More to the point, which is the methodology of search engines in KM, this module you talk about sounds like a winner to me. As does the whole concept of modular macros. As you mentioned before, modular reduces macro file size and app overhead, depending upon just how much "stuff" you care to load up.

I never thought the method for adding macros to earlier versions of KM was an elegant one, indeed it was a primitive cut and paste. It did do the job of allowing users to use as many search engines as they cared to load up though, but as you said, that very primitive method of updating made for confused posts from users who couldn't figure out how to add a search engine without posting requests for help here in the Forum. As you note, software development is evolutionary and striving for perfection is an ongoing process.

N

Quote
ndebord
More to the point, which is the methodology of search engines in KM, this module you talk about sounds like a winner to me. As does the whole concept of modular macros. As you mentioned before, modular reduces macro file size and app overhead, depending upon just how much "stuff" you care to load up.

I'm not sure about what you're talking now. Just to make sure: Have you tried km 1.1 beta now? (Released two days ago)

kko,

Yes, I tried KM 1.1 beta on my W98se system. Something hung and had to CAD to get out of it. Will try it again later on.

N