unRAID 5

Also, I should mention, this is not Ruby-on-Rails - rather Ruby on Sinatra.

 

If you don't mind me asking: Why Sinatra and not Rails?

 

I have very little experience with Ruby in general (started looking into it when I found the blurb on the web page) but from my admittedly little experience I have always heard mention of Rails and none for Sinatra.

If you use Slackware Current, they provide Ruby 1.9.1 patch243.

 

<slaps forehead> Of course!  For some reason, I was still thinking "slackware 13" = "slackware current" 

 

 

Yeah, it's a bit of a misnomer, but not really. Any Slackware # release is a snapshot capture in time. It only ever gets updates for security issues. You generally wouldn't install a MS WinOS and never ever update it, so why would Linux be any different?

Yes, BRiT, I know that.  It's that after a certain age, time starts playing tricks on you:

It felt like slackware 13 came out just last week.  Hense my looking for stuff in the wrong place.   

Wouldn't it be better to use something the majority of people know, and more importantly already know where the pitfalls are?

Ruby is not unknown by any stretch of the imagination.  Sinatra is a bit of a surprise, but there's surely a reason why Tom chose it.

 

Try not to hyperventilate until Tom documents something or releases an alpha.

 

I'm still holding out for Applesoft.

Tom will be giving us the hooks we've been asking for to allow add-ons to be cleanly started and stopped.  That, combined with an easy to integrate API makes me happy, regardless of the language/programming package he uses.

 

Tom already described his reasoning... ruby/sinatra fits with his existing structure (less re-write for him) and interests him as new technology.

 

It does not mean you cannot code an add-on in php.. or even, if you are so foolish, in "awk" or "shell"  . Your new plug-in simply might need a few lines of "ruby/Sinatra" to invoke it, and the install package for your add-on might need to have "php" installed.

 

Trust me, there will be plenty of examples of ruby/Sinatra once we get a bit of experience. 

 

For all those objecting because they don't know "ruby/Sinatra" neither do I.    I figure you'll get over the objections... eventually... and like me, learn enough to get the tasks done you would like to do in it.  As I said earlier, it is just a matter of if/where to put the semi-colons. 

 

Joe L.

Tom will be giving us the hooks we've been asking for to allow add-ons to be cleanly started and stopped.  That, combined with an easy to integrate API makes me happy, regardless of the language/programming package he uses.

 

Tom already described his reasoning... ruby/sinatra fits with his existing structure (less re-write for him) and interests him as new technology.

 

It does not mean you cannot code an add-on in php.. or even, if you are so foolish, in "awk" or "shell"  . Your new plug-in simply might need a few lines of "ruby/Sinatra" to invoke it, and the install package for your add-on might need to have "php" installed.

 

Trust me, there will be plenty of examples of ruby/Sinatra once we get a bit of experience.   

 

For all those objecting because they don't know "ruby/Sinatra" neither do I.    I figure you'll get over the objections... eventually... and like me, learn enough to get the tasks done you would like to do in it.   As I said earlier, it is just a matter of if/where to put the semi-colons. 

 

Joe L.

 

I agree with everything you have said.  I managed to get my hands on an older computer that I am going to try and set up for some dev testing when an alpha of 5.0 comes out.  I have been doing my own reading and searching on ruby/sinatra so I can at least have an idea of how it works.  Thankfully ruby is built into OS X on my Mac and it has been fairly straight forward to use.

 

Really looking forward to what is coming!!

....

 

For all those objecting because they don't know "ruby/Sinatra" neither do I.    I figure you'll get over the objections... eventually... and like me, learn enough to get the tasks done you would like to do in it.   As I said earlier, it is just a matter of if/where to put the semi-colons.  

 

 

Obviously your right:)

 

But equally you could say... "Lets all have a talk. We all speak English but lets use Esperanto. None of us have really know it but it easy to learn and is supposed to be good"

 

After all no one has actually said they are  experienced/expert in it so far, there alot to be said for choosing the devil everyone knows. Or put another way, Perl, PHP etc have been requested many many times.... ruby/Sinatra ... not once ever.

 

At the end of the day it makes almost no difference to me I just find it a curious choice.

 

My understanding supposition is that ruby will be used for extension management... not the API to unRAID.

 

What is unclear, is what the API to unRAID will work.... Http post?  Pseudo-device?  Inquiring minds want to know!

 

I too, have my bench box waiting to test an alpha of 5.0.

It seems like a few people are actually intending to use real hardware for 5.0 testing/development.  Am I the only one thinking virtualbox would be a great tool for it?

It seems like a few people are actually intending to use real hardware for 5.0 testing/development.  Am I the only one thinking virtualbox would be a great tool for it?

 

It gives me an excuse to use a machine I have sitting around, plus in this case I would prefer to mess with and actual machine instead of a virtual one.

At the end of the day it makes almost no difference to me I just find it a curious choice.

 

After doing a little homework, I got a hint on why the ruby/sinatra environment.

It seems to provide a concise framework and architecture.

I'm going to play around with it and sit tight to see what comes out of this one.

For anyone looking for info on Sinatra you can go to the Sinatra homepage or read this ebook.  There is quite a bit of information out there so it really is just a matter of waiting for 5.0 so we know what we have to work with.

Some cool links for your pleasure:  

 

try ruby! (in your browser)

http://tryruby.org/

 

Ruby book: Learn to Program, by Chris Pine

http://pine.fm/LearnToProgram/

 

Ruby Documentation Bundle (online)

http://ruby-doc.org/

 

Ruby Documentation Bundle (locally downloadable package)

http://ruby-doc.org/download/ruby-doc-bundle.zip

 

Sinatra: The Book

http://www.sinatrarb.com/book.html

 

RubyGems Manuals

http://docs.rubygems.org/shelf/index

 

Gems from Gemcutter

http://gems.rubyforge.org/

 

Rack Documentation

http://rack.rubyforge.org/doc/

 

The ruby package from slackware-current

ftp://slackware.osuosl.org/pub/slackware/slackware-current/slackware/d/ruby-1.9.1_p243-i486-4.txz

 

Joe L, Old is a relative term.  My professor for numerical methods in comp sci grad school was a fellow named Richard Hamming.  IIRC he had something to do with Bell Labs. He almost published my final exam in the Bell LAbs Journal (it was an unsolved problem) but he wasn't sure that I really solved it so we agreed I'd get an A for the course but he would not Publish it. That was 1962.

Joe L, Old is a relative term.  My professor for numerical methods in comp sci grad school was a fellow named Richard Hamming.  IIRC he had something to do with Bell Labs. He almost published my final exam in the Bell LAbs Journal (it was an unsolved problem) but he wasn't sure that I really solved it so we agreed I'd get an A for the course but he would not Publish it. That was 1962.

Hamming ... as in Hamming Codes?  I'm very familiar with them...  They were used in the first "computer memory" I worked on AT&T"  It was 47 bits wide, 2 20 bit half words, and 7 bits of Hamming and parity.

 

quoting the wikipedia

In telecommunications, a Hamming code is a linear error-correcting code  named after its inventor, Richard Hamming.

 

The memory could detect single or double bit errors.  It could detect AND correct single bit errors, on the fly.

 

I am impressed.  You are oooooooooooold.  (and smart too)

 

Joe L.

At the time he was my prof he was the director of Bell Labs and the editor of the Bell Labs Systems Journal (or whatever it was called then)

 

One of his lectures was on Hamming space and how the codes represent points in that space and how the Hamming distance measured the number of bits of corruption to a point (code) in hamming space before it could be mistaken for a different point in the space. In short, if the hamming distance of a code point in Hammimg space was two then at least three bits had to change in the code before it could be recovered. He, Hamming, explained it all in terms of multi dimensional geometry and made it easy to understand. Problem now is I forget too many details.

 

 

At the time he was my prof he was the director of Bell Labs and the editor of the Bell Labs Systems Journal (or whatever it was called then)

 

One of his lectures was on Hamming space and how the codes represent points in that space and how the Hamming distance measured the number of bits of corruption to a point (code) in hamming space before it could be mistaken for a different point in the space. In short, if the hamming distance of a code point in Hammimg space was two then at least three bits had to change in the code before it could be recovered. He, Hamming, explained it all in terms of multi dimensional geometry and made it easy to understand. Problem now is I forget too many details.

 

Sounds like an interesting concept.  I'll have to look into it too...  BTW, Barry, are you the same Barry Gordon from the Homeseer forums?  If so, glad to see you around here, you were an invaluable wealth of knowledge there...

 

G

At the time he was my prof he was the director of Bell Labs and the editor of the Bell Labs Systems Journal (or whatever it was called then)

 

One of his lectures was on Hamming space and how the codes represent points in that space and how the Hamming distance measured the number of bits of corruption to a point (code) in hamming space before it could be mistaken for a different point in the space. In short, if the hamming distance of a code point in Hammimg space was two then at least three bits had to change in the code before it could be recovered. He, Hamming, explained it all in terms of multi dimensional geometry and made it easy to understand. Problem now is I forget too many details.

You learned it in multi-dimensional geometry.  I learned it following bits and wires and transistor-diode logic on schematic drawings.  I had to repair the darned stuff, and to fix it you had to understand what it was supposed to do, and then figure out what it was not doing correctly. (We trouble-shot down at the bit level back then... pre IC chips, pre-motherboards... 16K of 47 bit wide memory was 4 feet wide and 7 feet tall and gave off about 2000 watts of heat (I guess it was not very "Green").  It was non-destructive read/write piggy-back-twister memory (a variation of core memory) 

 

I really did see the words "Forced Error Correction Mode On" print on the system console when I had both banks of redundant memory develop single bit faults at the same time at about 3 in the morning one night and the Hamming circuitry did its magic to supply the missing bit.  It was a long night.

 

I'm ooooooold too, and smart enough to admit I don't know-it-all. 

 

I once had a security audit program I wrote mentioned in Bell Labs Technical Journal. (My tiny bit of fame)  I was not mentioned, but my program was.  The author of an alternate security audit tool had written an article about his utility and was comparing it to "other notable tools" 

 

My audit tool, in use by the Bell Labs security team at that time, was one of the "other notable tools"

 

Joe L.

I don't know how I missed this reply.  We are of the same genration re Computers it seems. I also cut my teeth on playing with the bits.  No compilers, No cross assemblers, just paper tape, Giant boxes, oscilliscopes, all of that stuff.

 

For languages that end in "L" my favorite was APL (aka Iversons notation ny those who knew what it really was. It was a truly "elegant" language, that is it was really hard to read a program and ubnderstand it!!

Is there any roadmap / timeline for the release of unraid 5?

 

Is there any roadmap / timeline for the release of unraid 5?

 

None that are being shared with us. The last bit of info we had was shooting for a Christmas 2009 early alpha/beta release of unRAID 5.