Pl. suggest some beginning tips for new Merbists.

0 views
Skip to first unread message

MyMerb

unread,
Nov 21, 2008, 2:38:57 AM11/21/08
to merb
Hello,

Since Merb is new and so are we....can we expect some quick important
tips to move ahead from senior Merbists here...?

It would be nice to guide us on the points below.

(1) Best IDE currently available for Merb.
This is important since Netbeans is currently working on Merb support
for future.

(2) Best tutorials and slides to read ( Links are requested )

(3) Good Beginners projects available on GitHub to start with.

(4) A tutorial on working with Merb and console.

(5) Some Good Tips and links to visit frequently for learning more.

I hope this will help all fresh guys like me,willing to make a career
in Merb

Thanks

Michael Klishin

unread,
Nov 21, 2008, 2:43:27 AM11/21/08
to me...@googlegroups.com
2008/11/21 MyMerb <mym...@gmail.com>:

> Since Merb is new and so are we....can we expect some quick important
> tips to move ahead from senior Merbists here...?

Start writing an application to scratch some of your own itches (merb
tutorials website, for instance) instead of asking for "best
tutorials" on the mailing list.
You get better not when magic fairies come to give you "best IDE" and
"best tutorials" but when you read code and do actual work.
--
MK

MyMerb

unread,
Nov 21, 2008, 2:51:11 AM11/21/08
to merb
Hi MK,

Surprised to read a strange answer from your side.

When i am asking for tutorials to learn, you expect me to write
tutorials.

I am a fresh guy and yet not matured enough to create the projects
like you.

Don't force students to become the teacher the very first day. Its not
possible.

Thanks.

Julian Leviston

unread,
Nov 21, 2008, 3:04:46 AM11/21/08
to me...@googlegroups.com
Hey Dude,

I agree with MK.

He's not expecting you to write a tutorial.

He's saying "The best way to learn is to do, and ask questions when
you have problems, not to follow someone else's tutorials"

Having said this, there are tutorials on the web.

Use google.

Julian.

Doug Sparling

unread,
Nov 21, 2008, 3:11:05 AM11/21/08
to me...@googlegroups.com
You might dig through these:

http://www.rubyinside.com/44-merb-resources-1318.html

The book "The Merb Way" (the first four chapters are available in the Early Access Edition) should get you going too.

DAddYE

unread,
Nov 21, 2008, 4:39:30 AM11/21/08
to merb
Hello,

I don't apreciate so much reply on your question.

I think staff members need to much instead of standard replies.

MyMerb, in my opinion:

1) Best (is not an IDE) for mac is TextMate
2) Best tutorial for learn something about merb (there isn't yet a
lot) is: http://wiki.merbivore.com/example_apps/simple_app
3) Best docs now are on the wiki of merb
4) Best blog about merb are: http://merbist.com, http://merbunity.com/,
http://yehudakatz.com/

Remember that some two important things that introduce Merb:
DataMapper & RSpec

If you are interested in using it see some docs on their official
sites.

Then, on github there are some small project that use merb, all same
to be in alpha stage, but are very good to learn try for understand
some good tecniques.

Then, if you want to pay for learn there are some ways:

- Buy a peepcode Merb Draft: http://peepcode.com/products/meet-merb-pdf-draft
(It's a very good starting point)
- Buy the first 3 chatpers of Yehuda’s beta book on Merb:
http://www.manning.com/ivey/katz_meapch1.pdf

Then I see some where that core team made some courses for learn Merb,
but I forgot the link.

I hope you can walk with your leg now and I also hope members do
something more also for newbie or swichers than the standard replies:
"see the sources", "build an app from scrach".

Michael Klishin

unread,
Nov 21, 2008, 6:19:15 AM11/21/08
to me...@googlegroups.com
2008/11/21 MyMerb <mym...@gmail.com>:

> Don't force students to become the teacher the very first day.

The best way to learn is to try to teach.
I constantly see people moaning about tutorials, blah blah blah. You
have source, Merb is about 6K LOC,
merb is very similar to Rails in many aspects, Merb is a damn *small
web framework in Ruby*,
not an enterprise grade natural language processing system in C++.

If you want to learn it, it's usually more useful to start reading
sources and writing yourself a notes instead of asking people
for tutorials, best IDEs and building yourself a castle in the air.

IDEs and tutorials do not matter for learning at all, pick a simple
test editor and stop wasting your time on mailing lists.

> I am a fresh guy and yet not matured enough

Here you try to make a self-excuse. "I am weak, I am a fresh guy... I
need your help". You don't need any help from others.
Dig into source, write the best Merb tutorial out there, become a way
more skillful than Merb core team members (if you think me or Matt or
Yehuda or Dan are god likes, I have to disappoint you: we are mere
mortals just like you), then
apply your skills to change the world, become rich, famous or whatever
your dream is.

All I see now from you on the mailing list is constant asking for
things others should do for you: whether it is a NetBeans Merb module,
tutorials or something else. Dude, world around you changes when you
step up and write a NetBeans Merb module the world is lacking right
now,
or when you spend a week digging in the source to write a tutorial for
your friends, or when you submit a patch instead of calling Merb
contributors names.

It *is* that easy.
--
MK

DAddYE

unread,
Nov 21, 2008, 6:47:16 AM11/21/08
to merb
Mk,

you have reason, but is not applicable applicable to all.

Do you want that merb can be very popular?

If your reply is yes then there is some thing that you need to
remember.

Rails, become very very very very very famous with 3 screencast.

Why? Because unfortunately (for you) 80% of people, see something
interesting... in this case Merb, then:

1) search some tutorials
2) try a sample app
3) make some production application

If all is okey, submit bug, request, then read the source and finally
submit pach.

Read source require some time that 90% can't have... why?

For spent this time first a newbie/swicher need to have some answers
like:

1) Merb is good for me? (Seeing some tutorial can have part of the
answers)
2) Merb have a good support? (This is the place)
3) Merb is rock solid for big project? (For this rails have basecamp
example... Merb?)

I think when a person have this 3 simple responses can invest their
time. Otherwise we need to read all source of all interesting things
like sinatra, ramaze etc... but for people that work... is not
possible read all sources of all interesting things.

So, merb core team want bug report, pach, community is necessary
involve also newbie and do all for attract any one.

This because a newbie tomorrow can be a core team.

On 21 Nov, 12:19, "Michael Klishin" <michael.s.klis...@gmail.com>
wrote:

Michael Klishin

unread,
Nov 21, 2008, 7:21:05 AM11/21/08
to me...@googlegroups.com
2008/11/21 DAddYE <dad...@tin.it>:

> Do you want that merb can be very popular?

I want merb to save me from headaches I had with existing tools in the
Ruby world. I want it to be something Ruby community would not have
otherwise.
I don't care how popular it is, I don't feel for *software engineers*
who cannot figure out how a Ruby web framework of reasonable size
works. Merb is not rocket science, period.

I was honest with you. Now please don't tell me about why Rails is
popular, why something else is popular and how much of an asshole I
am. Use Merb if it fits your needs, make it better, help newbies if
you want to, pitch Merb to whoever you want. Just ignore me. I am an
idealistic asshole who thinks that 6K LOC in Ruby is not that scary
and people could do better scratching their own itches.
--
MK

MyMerb

unread,
Nov 21, 2008, 9:50:39 AM11/21/08
to merb
Dear MK,

Your advice is nice, but it does not fit everyone.

You can say many words based on your experience. In fact with your
experience, you should be helping newbies, not cursing them.

You are expecting a child to run, when he does not even know how to
walk.

You are keeping a track on my threads for requests and blaming me. In
fact you should think the other way. I am constantly trying to learn
from the senior ones over here.

I cannot leave my hope, if some one like you does not support here.
There are plenty ones around to guide me.

Do not forget, even MERB has taken 2 years to develop, with an
inspiration from Rails. Did you ask them, how much pain they had
taken in developing.

Even the Rails Core Team and Merb Core team were newbies, when they
started.

If you cant help, its Ok. but don't discourage them from asking a
genuine help from others.

Thanks

Julian Leviston

unread,
Nov 21, 2008, 10:15:22 AM11/21/08
to me...@googlegroups.com
Let's be very clear.

You have asked a question.

You got an answer. It wasn't the one you wanted, but it was an answer!

Now, can you listen to the answer?

The answer is to build something yourself, no matter how small. Choose
one thing you're interested in, and don't follow others.

If you are really interested in learning from people more senior to
you, take Michael Klishin's advice, and TRY TO BUILD SOMETHING VERY
SMALL YOURSELF AND ASK QUESTIONS WHEN YOU GET STUCK.

This is, quite honestly, the best advice anyone could give you.

If you want to learn Japanese, you can do courses, and you can ask
people (in English) how to talk in Japanese, but it's not going to
help you one WHIT at actually having functional skill in the language.

To learn the language, you have to TRY TO USE IT.

It may SEEM hard, but it's not. Set yourself a VERY small goal and try
to accomplish it.

Now, perhaps, the questions you're having trouble with is where to
begin. Well and good, but you need some REAL PROBLEM - some real need
of YOUR OWN to drive this endeavour for yourself.

Julian.

Michael Klishin

unread,
Nov 21, 2008, 10:36:13 AM11/21/08
to me...@googlegroups.com
2008/11/21 MyMerb <mym...@gmail.com>:

>
> Dear MK,
>
> Your advice is nice, but it does not fit everyone.

It does, you did not try (yet).
--
MK

Michael Klishin

unread,
Nov 21, 2008, 10:56:47 AM11/21/08
to me...@googlegroups.com
2008/11/21 Julian Leviston <jul...@leviston.net>:

> If you want to learn Japanese, you can do courses, and you can ask
> people (in English) how to talk in Japanese, but it's not going to
> help you one WHIT at actually having functional skill in the language.
>
> To learn the language, you have to TRY TO USE IT.

I know everyone is sick of me here, but learning languages is a
perfect example of how people really learn and accomplish something.
Keep in mind learning a programming languages usually 1000 times
easier thing to do compared to natural languages.

I study Italian. I have no idea why, I just like learning languages.
How I do it, you ask? I listen to some radio shows in Italian,
understand nothing,
read a few lessons in a textbook, next day I listen to it again. After
a week I started to understand certain sentences. I feel myself
extremely stupid every time I listen to the program. But I also
understand that if I get all the fancy language coursed you can find
today,
if I start studying with the best teachers I can find in my area, but
then *do not really push hard enough*, I won't learn a thing. So fancy
text books and other fancy things are ignored intentionally.

In contrast, if I continue using one textbook and listen to the same
radio show for months and months, one day I'll be understanding it
all. Key word here is *listen for months and months*, *doing*, this is
what most people do not do. Instead, they build themselves a castle in
the air, trying to imagine how *awesome* they will be if they ask
"senior people" and get all the fancy IDEs/textbooks/etc. They collect
gazillions of books and links they never read, install 20 IDEs they
never use, try to follow "best git workflows" and "best git practices"
where Mercurial with a single branch will do, and so forth.
Don't do it. Don't fool yourself.

So if you ask my heart advice without bitching about newbies, here it
goes. Forget about fancy IDEs, "best tutorials", "best operating
system to develop on". Pick first text editor you can find (on Linux
and Mac I'd recommend Vim since it's there by default, and I am a die
hard Emacs junkie: it does not matter when you learn), imagine what
would be a fun thing to do for you and get lost. Never come back to
any mailing lists until your goal is completed and you are ready to
share your experience with others.

This is how Merb, Mongrel and a lot of other projects started.
--
MK

Yehuda Katz

unread,
Nov 21, 2008, 11:44:24 AM11/21/08
to me...@googlegroups.com
Hey guys,

I'm sorry I wasn't here earlier in this conversation. We absolutely need tutorials, examples, etc. At the moment, we're pretty thin on a lot of what you're asking for, but let me see if I can answer some of your initial questions:

(1) Best IDE currently available for Merb. This is important since Netbeans is currently working on Merb support for future.

At the moment, pretty much any IDE available with Ruby support will also support Merb. This includes NetBeans, IntelliJ, Aptana, etc. However, no editor except Textmate currently has Merb-specific features. A lot of Ruby developers find Textmate to be a perfect compromise between a raw text editor and a full-fledged IDE; obviously, your mileage may vary. I'd definitely keep an eye on NetBeans :)
 
(2) Best tutorials and slides to read ( Links are requested )

The videos for all of MerbCamp are available: http://www.merbcamp.com/video. The getting up and running talk is a bit outdated (it actually was outdated even when he gave it), but there are some very good tutorials on the wiki under "How to": http://wiki.merbivore.com/.
 
(3) Good Beginners projects available on GitHub to start with.

One of the best examples I've seen so far is http://github.com/nanodeath/merb_example_google_rss/tree/master, an example app that takes in a Google search and outputs an RSS feed using Haml.
 
(4) A tutorial on working with Merb and console.

Hmmm... I don't know if we have anything like that. What specifically are you looking for?
 
(5) Some Good Tips and links to visit frequently for learning more.

I'd keep an eye on the Merbist (http://merbist.com) and my blog (http://www.yehudakatz.com). You can also check out http://planet.merbivore.com/, which aggregates the blogs of some of the top merbivores around. And of course, keep an eye on the wiki!

Regarding Michael's comments, I'd say his suggestions are more appropriate for people wanting tips on how to become a kick-ass programmer in the abstract. The only way to become legendary is to write an application that scratches your own itch and push the boundaries of your existing skills. I always say, "look for a project that seems out-of-reach impossible, then do that."

But you obviously can't get to that point before you get up and running; I hope I've been somewhat helpful on that front. I remember when I first started with Rails, and I wasn't very sympathetic to people who told me to stop whining and buck it up, so I can imagine that you feel the same way. Again, hope that my answers have been helpful :)

-- Yehuda

--
Yehuda Katz
Developer | Engine Yard
(ph) 718.877.1325

Michael Klishin

unread,
Nov 21, 2008, 12:31:27 PM11/21/08
to me...@googlegroups.com
2008/11/21 Yehuda Katz <wyc...@gmail.com>:

> I
> always say, "look for a project that seems out-of-reach impossible, then do
> that."

If someone starting with Merb needs a few ideas what to work on,
people often ask me why Merb wiki is not written in Merb.
And though I do not care what it is written in, Collective could be
improved. Grab the source at github [1], see what moinmoin [2] can do,
clone it's Mercurial repository [3]
to see how they organize domain model (no need to read through all of
Python code there, just figure out what their models are), then start
working on Collective.

And of course if some specific question arises ("I try to do a diff
between two versions, how would you do it?"), I am sure people on this
list will be glad help you.

[1] git clone git://github.com/meekish/collective.git
[2] http://moinmoin.wikiwikiweb.de
[3] hg clone http://hg.moinmo.in/moin/1.9 moin-1.9
--
MK

Matt Aimonetti

unread,
Nov 21, 2008, 12:51:30 PM11/21/08
to me...@googlegroups.com
Sorry I'm at Qcon and didn't really check my emails lately :(  Here are my answers to your questions:


(1) Best IDE currently available for Merb.
This is important since Netbeans is currently working on Merb support
for future.

IDE choice really depends on your needs/taste. TextMate/e-editor work fine with Merb. Yehuda and I are working with the NetBeans team to offer Merb integration so people liking NetBeans can use. Most editors supporting Ruby will work fine with Merb, we don't have one recommended IDE.


(2) Best tutorials and slides to read ( Links are requested )

We are working on making that better. For the moment you have examples on the wiki:

http://wiki.merbivore.com
a nice peepcode from peepcode.com with a full example app (twitter client)


(3) Good Beginners projects available on GitHub to start with.

http://github.com/mattetti/simple_merb_example_app/tree/master

It's not much but I hope to be able to write more example. Hopefully people will also write more examples for people to learn Merb.


(4) A tutorial on working with Merb and console.

The wiki covers some examples, the console is really simple you can just access it by doing:

 merb -i

In your command line.

The 3 Merb books might help covering this kind of information if you don't have time to try things or wait until we get more information in the wiki.


(5) Some Good Tips and links to visit frequently for learning more.

Look at the code on github, check my blog: http://merbist.com, and other blogs: http://merbunity.com and http://yehudakatz.com


What would be nice is that if you learn something new you could share with people using the wiki. This way we could have a better documentation.

- Matt

Rich Morin

unread,
Nov 21, 2008, 12:56:37 PM11/21/08
to me...@googlegroups.com
I've been working with computers since 1970, and I haven't been
sticking with a single language, tool set, etc. So, I've had a
number of opportunities to be a newby. (The only folks who are
never newbies are the ones who never try anything new. :-)

My take, FWIW, is as follows:

* Different resources help folks to learn in different ways.
Reading a book is very different from watching a screencast.
Both are different from reading and/or writing code. I use
ALL of these techniques (rinse, repeat :-) as I progress.

* There is NO substitute for getting your hands into the code.
I know; it's scary. But at some point, you have to jump on
the bicycle, fall off, and try again. Accept your scrapes
and bruises as badges of honor.

* Merb is currently a moving target, though the 1.0 "line in
the sand" will be an enormous help. Expect rapid changes
over (at least) the coming year. Expect Merbists to find
and/or build several new plugins, slices, tools, etc.

I'm pretty comfortable (though not all that happy :-) with Rails,
so my approach has been to compare Merb to Rails. What parts are
different? How does this affect the way I'd do things. Etc.

Finally, I try not to get too upset by either nattering newbies
or gratuitous grumps. I play both roles myself, at times, so
I'd best be tolerant of the behavior in others.

-r
--
http://www.cfcl.com/rdm Rich Morin
http://www.cfcl.com/rdm/resume r...@cfcl.com
http://www.cfcl.com/rdm/weblog +1 650-873-7841

Technical editing and writing, programming, and web development

grantmichaels

unread,
Nov 21, 2008, 7:18:53 PM11/21/08
to merb
MK: I needed your advice months ago ... In fact, I just realized
within the past few weeks how unproductive my time usage has been for
the past 10 months. I've ended up well read w/ little practical
experience and for the most part have come up fruitless ... well,
until earlier this week when I wiped and restarted my host and removed
most all of the distractions I'd accumulated. Your message was served
up pretty dry, but it might actually be the most effective solution to
his problem, at least it would have helped me not get to go ankle ->
knee deep in a dozen language/framework combinations on 2 platforms
and a half dozen OS's trying to find the 'perfect fit' (which seems to
not exist anyways) ...

Step 1 - Acknowledgement that there is a problem ...

Step 2 - Resolution to take smaller bites ...

Step
.
.
.

-grantmichaels

On Nov 21, 10:36 am, "Michael Klishin" <michael.s.klis...@gmail.com>
wrote:

jonuts

unread,
Nov 22, 2008, 11:29:46 AM11/22/08
to merb
On Nov 21, 6:44 pm, "Yehuda Katz" <wyc...@gmail.com> wrote:
>
> (1) Best IDE currently available for Merb. This is important since Netbeans
>
> > is currently working on Merb support for future.
>
> At the moment, pretty much any IDE available with Ruby support will also
> support Merb. This includes NetBeans, IntelliJ, Aptana, etc. However, no
> editor except Textmate currently has Merb-specific features.

I would just like to take this moment to pimp merb.vim (http://
github.com/nono/merb.vim). I stumbled upon it recently, and can't
begin to describe the pleasure of having my DM 'property' values
colorized!
Reply all
Reply to author
Forward
0 new messages