Digitization of old CD's

Since i haven't played a CD in some years, i finally got around to getting rid of the dusty cd cabinet in the upstairs 'media room' sometimes called living room. As i boxed them up i realized there are some songs i don't have digital copies of so i will have to either spend time searching and downloading or digitizing them myself. Most of the stuff i have is eclectic so it is unlikely to be in an online music store.

Now that we are using Connect360 on the mac pro for all the existing MP3 files and Tversity on the windows server for all the FLAC's and movie files, there is no need to keep audio CD from long term storage. We will not be converting any audio cd's to MP3 but rather to the higher resolution FLAC as digital masters and then downconverting to MP3 as needed.

The problem is that i want to use the mac for ripping the CD's but i don't want to use iTunes for that. I want to rip to high resolution FLAC's and then convert to MP3's if needed. First we take a look at Max (http://sbooth.org/Max/) for OSX. While this is still beta (less than version 1.0) it has a lot of nice features. We set the output as FLAC to a folder on the external drive and insert and an audio CD. It gets the metadata from the net, rips and converts the files and ejects the disc, all in 3 minutes. Excellent!

Wonder if it supports multiple drives, i have a couple of Firewire external cd-rom drives, a plextor and a lacie i could hook up. All you would have to do is to feed it a CD every few minutes.

Greenstone Z39.50 lookup to Library of Congress

In Greenstone 2.83 for Windows click the Download tab and select Z3950. Type in the following settings:

HOST: z3950.loc.gov
PORT: 7090
DATABASE: voyager
FIND: @attr 1=4 "SQL"

This sets up a connection to the Library of Congress Z39.50 server and does a search for books with SQL in the title.

Tick the max records checkbox and set the number to 5.
Click Download to start the retrieval

The search log indicates it completes successfully but no file of marc records is found in Application Data folder:

from Yaz log:

Opening connection to z3950.loc.gov:7090
Access database: "voyager"
Searching for keyword: "@attr 1=4 SQL "
<>
Yaz is Gathering records: 1 - 50
<>
Saving records to "\files3\Faculty$\users\gnickers\Application Data\Greenstone\GLI\cache\z3950.loc.gov\voyager_@attr 1=4 SQL _50.marc"
Closing connection...

The log file was written in:

U:\Application Data\Greenstone\GLI\log\download-Z3950994507293.log

but no .marc file. I tested with 500 records and 50 records and 5 records. I created z3950 folder manually and re-ran query. No joy. The Greenstone documentation notes:

"You can view the downloaded MARC files on the Gather panel. On the left-hand side of the panel, double click the Downloaded Files folder to expand its content. The subfolders are named by the Z39.50 server url. The MARC files are named as the combination of database name, query, and max_records if max_records is specified. These MARC files are physically stored in a temporary cache directory. "

This could be a Greenstone bug, further investigation reveals 3 PERL errors. The c:\program files\greenstone\perlib\downloaders\Z3950Download.pm file at:

Line 146
Line 156
Line 159

Next we replaced the voyager database label with

LCDB

Which i got from the loc page. In fact i replace the greenstone settings, which seem quite old with:

HOST: lx2.loc.gov
PORT: 210
DATABASE: LCDB
FIND: @attr1=4 "SQL"
MAX RECORDS: 5

which retrieved 5 records but none were saved. I changed the MODE setting from Librarian to Expert and re-ran the query. No joy. We then removed the max records setting and got 500 hits. Nothing saved.

I also tried the OAI harvester function, although this is not supposed to work on the local version, only on the server install.

Select Download, OAI and enter the following settings:

SOURCE URL: http://memory.loc.gov/cgi-bin/oai2_0
RESTRICT to SET: cwp
MAX RECORDS: 5

Click DOWNLOAD and see what happens. 0 of 5 files are retrieved and the log shows:

Parsing OAI identifiers.....
<>
Gathering OAI record with ID oai:lcoa1.loc.gov:loc.pnp/cph.3a00275.....
Unable to save oai metadata record: No such file or directory

Looks like it kind of worked except there was no place to save it or no file to save. An examination of the perl code might help.

I also tried the SRU function using the information supplied in the Greenstone documentation but we could not get a connection. The log shows:

Opening connection to http://z3950.loc.gov :7090
Cannot connect to http://z3950.loc.gov :7090

We tried the LOC SRW server at:

http://lx2.loc.gov:210/LCDB? 

But Greenstone gave error messages in trying to connect.
For details on the LOC server see: http://www.loc.gov/z3950/lcserver.html

Would be very interested in seeing an example or hearing from someone who has this working!

Skype

Liam has been chatting with his chum via ye olde pots but it is long distance. So i was thinking of Skype as a solution that would put off getting him a cell phone for a bit...

He has an iMac so i signed him up for Skype and installed it. Worked fine with the iSight and his mic, a logitech usb upgrade from the built-in mic. I restricted his calls to people in his contact list and filled in the list with me to start.

We tested it out this weekend with us calling upstairs/downstairs and he thought it was pretty cool. The only problem was this is the 1st week of term and the Rogers network can't keep up - the connection is up and down like a yoyo. My theory is their DNS server gets overloaded and things time out but who knows for sure? Anyway it is was moderately successful as a test - next step is to get his local and far away chums signed up and let them have have it.

And i have to add it to my linux laptop too...just need to get the damn dell webcam working. Vendors who build equipment using parts such as wireless and cams that are known not to work with linux is beyond my comprehension - i would love to see 'linux-certified' stickers on equipment - because i would pay for that!

GPT Protective Partitions and Windows XP

I upgraded the hard disks in the mac pro to 1 gig models and put the 500mb old drives to use, one as a time machine for the macbook, another in the media center and the third as a 'clone' drive to backup the last XP box called Gamera.

The idea is to automatically clone the xp system disk in case of failure. The document files are stored on the network but re-installing windows and it's application is not a welcome chore.

The problem was that XP could not format the mac osx drive. It listed it as a GPT Protected Partition but could not access it. I booted the Gparted liveCD which listed the drive and seemed to allow me to change the partition to NTFS but when i rebooted into XP it was still locked into GPT.

The solution was to drop into the windows terminal or CLI by running cmd. At the command prompt type:

diskpart

to run the disk partition utility. You can now list your disks, select a disk and clean it up. In this case we see there are 4 disks, disk 0 being the XP system disk and disk 1 being the osx disk.

DISKPART> list disk

Disk ### Status Size Free Dyn Gpt
-------- ---------- ------- ------- --- ---
Disk 0 Online 466 GB 0 B
Disk 1 Online 466 GB 0 B
Disk 2 Online 699 GB 0 B
Disk 3 Online 932 GB 0 B

DISKPART> select Disk 1

Disk 1 is now the selected disk.

DISKPART> clean

DiskPart succeeded in cleaning the disk.

Once the disk is cleaned, then run control panel, administrative tools. computer management, disk management.

This brings up the disk initialization wizard which allows you to select disk 1 and to format it as NTFS. The disk now shows up correctly as sys (G:) 465.76 gb NTFS.

The next step is to find a windows version of the great osx utility - carbon copy cloner. The requirements are:

- creates an exact copy of the system disk including boot data and permissions
- updates the duplicate disk automatically on a schedule
- open source or non-proprietary


I think DriveImage might work but first we download XXclone. It's pretty simple, you can set the backup drive to be bootable and then copy all the files. It also has an incremental copy option.

Next we run DriveImage XML which is free for personal use. It has a backup option but also a drive to drive copy. We decide to test copying c: to g: which will take some time...

The Koha Laptop Project

The idea is to build a portable 'server' that i can take to class to demo stuff and to bring to small labs so students can use it like a server. How well the second idea works out remains to be seen. The first step was to find a laptop that works with Linux and is cheap but powerful enough.

First attempt involved an HP dv6000 but i could never get the wireless to work. So that went to the flea market. I was tempted to use the dell D400 in the kitchen but it is underpowered for a server, fine as a firefox platform. I checked dell's canadian site but couldn't find any with Ubuntu anymore so i looked at the list of dell machines known to run linux.

The flea market yielded a very cheap Dell Inspiron 6400. Cheap because there is no battery, the Vista OS wouldn't boot, and it was missing an ALT key but those are not needed.

The machine is decently powered with 2 gig ram, 120 gig hard drive and a P4 duo core cpu. The seller told me it had intel pro wireless so it seemed a good choice.

I installed 9 but the wireless (the dreaded broadcomm 43xx - the seller was wrong) did not work so i upgraded to version 10 and it was able to see my WAP's.

Next - setup Koha

Greenstone - Cannot import PDF files

The problem is when adobe comes out with a new version of the file format it is incompatible with older versions, for example some new pdf's cannot be imported into Greenstone. This is noted in the log:

"PDF version 1.6 -- xpdf supports version 1.4"

The solution would be to convert existing pdf's to an older version such as 1.4 or earlier. If the pdf's were produced from some other files using a program like Open Office should produce pdf's that Greenstone can use.

http://www.topshareware.com/pdf-1.6-to-pdf-1.4/downloads/1.htm lists PDF Version Converter 1.0, a program that converts Convert PDF file between different versions for compatibility purposes. I've not tested this. Should be other programs or even a web site that performs a similar function, a program that works in batch mode would be preferable.

Koha on the HP 6000

This morning i decided as an experiment to change the old windows xp laptop, an hp 6000 into a dual-boot machine with one half being linux running the koha library system. The laptop sits in the corner unused 90% of the time anyway. I no longer have any interest in windows, xp is useful occasionally for legacy applications but using vista or vista 2, er windows 7 would be huge steps backward in functionality and usability from my macpro labtop or the ubunutu laptop.

The first step was to divide the hard drive into a windows NTFS partition and an Linux ext3 partition. There are commercial programs to do this, i used Partition Magic for years and found it a good, reliable application but it's not installed and the version i own is old. So we download a bootable Gparted cd image and boot it from the optical drive.

The Gparted live CD contains a ultra minimal Linux environment, complete with a sparse GUI. The gparted tool runs and shows you the existing hard drive(s) and their partitions. The first step is to shrink the existing NTFS partition to 60 gig. Plenty of space for XP as I'm not likely to add any new windows apps. Are there new windows apps? Name a 'killer' application that you would purchase windows for. No one buys a OS to run an OS, it's a platform to run applications that you use to do things. The OS is an enabler that should just work. I can't think of anything i want to run on windows besides the occasional use of Microsoft office for work.

Anything i used to do on windows i now do in Linux, at a lot less cost. I bought DOS 2.2, 3.0, 4, 5, 6.22 etc and Windows 2.11, 3.0, 3.11, 95,98,2000, xp, vista and many, many applications over the years. The only software purchases i've done lately are OSX 10.3, 10.4, 10.5, 10.6 that came with various macs, and iLife/iWork that did not (and some small online purchases from app store, and web sites like Tversity).

So now about 90% of the things i do are done in open source applications or on a mac. The last reason for using windows is that workplaces are still windows-centric. One would think Microsoft would concentrate on making those things work better (think Sharepoint) instead of trying to dominate the 'youth market' with zune, xin?, xbox etc. It's like your dad trying to be cool by growing a ponytail and wearing a Hawaiian shirt...let's face it, Windows is your Dad's operating system and Microsoft is the Dad 'n Lad brand of OS, and no stupid marketing campaign can change that fact so Microsoft - stop trying to be cool and accept you are a middle aged geezer in a white shirt and tie who works in big government and industry.

Anyway, now that i think about it why do i need a windows laptop anyway? The machine is too slow to run Vista (1.8 ghz dual core, 2 gig of ram) acceptably and there is no windows programs i really need on a laptop. So we decide to go the whole hog and boot up with the Koha Live CD and install, using the entire 120gig hd and wiping out Windows. Bye bye. We are now down to one lonely windows desktop which will probably go this year.

The install goes fine, and it picks up the wired connection and offers to install the proprietary drivers for the broadcom wireless and nvidia graphics. Next we want to upgrade from 9 to 10.4 but that does a fair bit of downloading so to save the internet bills we might hump the machine to school tomorrow. This would give us a fairly decent linux box with a fully functioning version of Koha. Sweet. There are a ton of apps on the dual xeon box so we want those too! But for now we grab the ubuntu-resticted-extras from the synaptic package manager. At a download rate of 1003kb/sec this does not take long.