Book about multivalue databases theory

[Rodrigo Zanatta Silva]

Hi. My company use the PICK/D3 database. I am not a programer, but I want to understand the MultiValue database. All I studied in my life was the normal relational database. Is there a book to understand this kind of database? What I can read to understand this kind of paradigm? 

[Dawn Wolthuis]

For a book, check out jes.com.

I also invite you to start with my little flash card trilogy. It looks better on actual cards as I have not optimized it this way but check it out the slides at http://www.tincat-group.com/mv/

There are some other good resources, such as a getting started guide I saw at one point. Several folks teach classes.  Best.  --Dawn

Sent from my iPad

On Jan 11, 2018, at 2:17 PM, Rodrigo Zanatta Silva <rodrigoza...@gmail.com> wrote:

Hi. My company use the PICK/D3 database. I am not a programer, but I want to understand the MultiValue database. All I studied in my life was the normal relational database. Is there a book to understand this kind of database? What I can read to understand this kind of paradigm? 

--
You received this message because you are subscribed to
the "Pick and MultiValue Databases" group.
To post, email to: mvd...@googlegroups.com
To unsubscribe, email to: mvdbms+un...@googlegroups.com
For more options, visit http://groups.google.com/group/mvdbms

[Cleber Vieira Baptista]

Install a personal version of OpenQM and explore the manual. I think this is the best way to learn how to work with MultiValue.

http://downloads.openqm.com/help/index.html

http://www.openqm.com/

Boa sorte,

Cleber

2018-01-11 18:17 GMT-02:00 Rodrigo Zanatta Silva <rodrigoza...@gmail.com>:

Hi. My company use the PICK/D3 database. I am not a programer, but I want to understand the MultiValue database. All I studied in my life was the normal relational database. Is there a book to understand this kind of database? What I can read to understand this kind of paradigm? 

--

You received this message because you are subscribed to
the "Pick and MultiValue Databases" group.
To post, email to: mvd...@googlegroups.com

To unsubscribe, email to: mvdbms+unsubscribe@googlegroups.com

For more options, visit http://groups.google.com/group/mvdbms

--

Cleber Vieira Baptista

+55

 61 8441-1663 Oi

 

[Dawn Wolthuis]

I forgot to mention perhaps a good place to start for someone who has worked with relational databases is here http://www.tincat-group.com/mewsings/2006/01/is-codd-dead.html

I will be a guest speaker in a database course next month and have a few tidbits, in case it is useful.

The MultiValue data model is the key to understanding a MultiValue database. In fact, I use SQL Server as a data store for multivalue data. Most MV DBMS products have hash table implementations with ONgroup (implementation on SQL Server, Oracle, or a DBMS of choice) and InterSystems Cache' (b-tree implementation in the Cache' DBMS) being notable exceptions.

This data model looks a bit like XML or JSON with the tags removed and funky characters used as delimiters for records, fields, and values instead.

A decade ago we had to apologize for MV data not being in first normal form, however, the industry has moved back to non-1NF, also known as NF2 (Non-first Normal Form) data structures with JSON and XML in particular.

This is an illustration of 2 of Codd's rules that the MultiValue data model, along with other NoSQL data models (for document databases) breaks

t

This is a good thing. Those rules were highly unnecessary. However, there is one really cool thing about first normal form. It lends itself to first order predicate logic, so that, in theory, anything one would need to get in a result set could be retrieved using SQL. With MultiValue, the queries, in general, use a portal approach, with the query choosing a single file for the query and then reaching from there into other files as needed to collect the data.

Here is another slide from one of my ppt decks

I have a lot more on the history and data modeling theory for relational and NoSQL/MV data models, and I know that in this forum collectively there is a ton of knowledge, so feel free to ask questions.

--Dawn

Dawn M. Wolthuis

Take and give some delight today

On Thu, Jan 11, 2018 at 2:17 PM, Rodrigo Zanatta Silva <rodrigoza...@gmail.com> wrote:

Hi. My company use the PICK/D3 database. I am not a programer, but I want to understand the MultiValue database. All I studied in my life was the normal relational database. Is there a book to understand this kind of database? What I can read to understand this kind of paradigm? 

--

[Wol's lists]

Also look at Pickwiki (www.pickwiki.com) (although it seems to down
right now - I can't access it ...)

The best way I like to look at Pick is to think of it as "relational on
steroids" although I'd rather classify relational as "break everything
down into its constituent atoms" - and you can't understand how a car
works by looking at the individual iron atoms that make up the sheet
steel ...

Pick is an EAR database (entity, attribute, relationship) database. All
the rows that relate to an entity are stored in a single "row", but
because it's a dynamic array, it can store a row inside a cell. So
instead of being a one-dimensional array, a row is now a n-dimensional
array.

This results in the quote from the Pick FAQ - "SQL optimises the easy
task of finding data in RAM, Pick optimises the hard task of getting it
into RAM in the first place". Each "row" stores all the data associated
with an instance of an entity, including foreign keys of related
entities. So firstly, accessing any attribute will retrieve all
associated attributes at the same time, and drilling down is easy and
fast because the entity contains the necessary keys for any related
entities.

And because Pick maps its internal structure closely to the real-world
structures, it's easy to understand - unlike relational as I said above
where everything has to be broken down into excessively intimate detail.

Cheers,
Wol

[Rodrigo Zanatta Silva]

I am a programmer, but not the one that work with this kind of database.

[Martin Phillips]

Not a book but take a look at http://www.openqm.com/docs/mv-intro.pdf. Although this is on the OpenQM web site, it is equally applicable to all multivalue products.

 

 

Martin Phillips
Ladybridge Systems Ltd
17b Coldstream Lane, Hardingstone, Northampton NN4 6DB, England
+44 (0)1604-709200

--

You received this message because you are subscribed to
the "Pick and MultiValue Databases" group.
To post, email to: mvd...@googlegroups.com

To unsubscribe, email to: mvdbms+un...@googlegroups.com

[Tony Gravagno]

Coming soon... an entire website dedicated to questions like this, and much more...
T

[jes]

On Thursday, January 11, 2018 at 3:44:33 PM UTC-5, Rodrigo Zanatta Silva wrote:

Hi. My company use the PICK/D3 database. I am not a programer, but I want to understand the MultiValue database. All I studied in my life was the normal relational database. Is there a book to understand this kind of database? What I can read to understand this kind of paradigm? 

If by "understand" you mean how to use, there are still books in circulation if not in print. 

There's a list of known or suspected Pick titles on my site, jes.com, along with my free "Pick/BASIC: A Programmer's Guide".

If you actually mean the theory underlying the platform, no such works are known to exist,

and the handful of people qualified to write them are currently disinclined to take the

time to write it all down. I know. We speak of it often. It would have to be approached as a 

labor of love for digital historians rather than for commercial success. Pretty much

like the platform itself.

If you want to skip the book search and get up to speed in 10 to 20 hours, I have an option for that.

j.

[Wols Lists]

On 17/01/18 14:32, jes wrote:

>
> If you actually mean the theory underlying the platform, no such works
> are known to exist,
> and the handful of people qualified to write them are currently
> disinclined to take the
> time to write it all down. I know. We speak of it often. It would have
> to be approached as a
> labor of love for digital historians rather than for commercial success.
> Pretty much
> like the platform itself.
>

Pickwiki is back again ... www.pickwiki.com. Take a gander round that -
a couple of pages of particular interest are

http://www.pickwiki.com/index.php/PhilosophyOfPick
http://www.pickwiki.com/index.php/MVDefinition

and both of these pages are accessed from

http://www.pickwiki.com/index.php/MultiValuedDatabases

Cheers,
Wol

[Scott Johnston]

I don't know maybe its just me but, it's very simple.

I look at Standard DB's like walls built with hallow legos.  You build them, put data in them but, if you need something new, you have disassemble them and rebuild,

The MV DB, I look at like a sheet of endless loose leaf paper and the field name goes in the margin .  You can build up and down without limits (I'm not talking about addressing inefficiencies here) and if you want to point to something on another sheet of paper (file) you can still write in the margins (a new dictionary) item. To make it even easier to find, you write an index file just like a book.

It couldn't be simpler.  

Now, when it comes to creating and writing efficient database structures, that's an art! I've learned that no two companies needs are the same and the MVDBs are so easy to change, that a "solution" can be developed very quickly.  But there's the rub!  That practice can quickly get out of hand. No amount of study can help you with controlling that. One has to be disciplined in how they implement changes.  

Someone wrote about downloading OpenQM.  Zumasys offers eval copies of jBase *the best MV - EVER!" and Rocket offers Unidata and UniVerse.  All have one year evel licences.  

My question to you: Why are you interested in MV at this point?  

On Thursday, January 11, 2018 at 3:44:33 PM UTC-5, Rodrigo Zanatta Silva wrote:

[Scott Johnston]

HOLD IT =  You wrote, " I am not a programer, but I want to understand the MultiValue database". 

So are you really BUTTER or a butter substitute? 

[Will Johnson]

https://www.amazon.com/Pick-Basic-Programmers-Guide-library/dp/0830628452/ref=sr_1_1?ie=UTF8&qid=1516733262&sr=8-1&keywords=jonathan+sisk

[Will Johnson]

Looks like there's one here for six bucks

https://www.amazon.com/Exploring-Pick-Operating-System-Jonathan/dp/0810462869/ref=pd_sim_14_4?_encoding=UTF8&pd_rd_i=0810462869&pd_rd_r=4ZS7KRGPGAQ3VGVVYCW6&pd_rd_w=yLWzu&pd_rd_wg=Q5nBv&psc=1&refRID=4ZS7KRGPGAQ3VGVVYCW6



On Tuesday, January 23, 2018 at 10:48:27 AM UTC-8, Will Johnson wrote:

https://www.amazon.com/Pick-Basic-Programmers-Guide-library/dp/0830628452/ref=sr_1_1?ie=UTF8&qid=1516733262&sr=8-1&keywords=jonathan+sisk

[Tony Gravagno]

PickWiki never went away.
Did it?

[Wols Lists]

On 25/01/18 19:50, Tony Gravagno wrote:
> PickWiki never went away.
> Did it?

See my previous post in this thread - there was a problem with it and I
couldn't access it. May have been just me, may have been a glitch at
their end.

>
> On Wednesday, January 17, 2018 Wol wrote:
>
>

> Pickwiki is back again ... www.pickwiki.com <http://www.pickwiki.com>.
>
Cheers,
Wol

[chandru murthi]

Reflections on the History of Computing:

Preserving Memories and Sharing Stories

 

edited by Arthur Tatnall

EVALUATING DATABASE MANAGEMENT SYSTEMS:

A FRAMEWORK AND APPLICATION  TO THE VETERAN'S

ADMINISTRATION  HOSPITAL

By MOHAMMAD DADASHZADEH S.B.,

Massachusetts Institute of Technology

(Google othb) have  interesting stuff on Pick.

[Ian McGowan]

It was down briefly, but is back again.  There are a couple of technical articles that I refer to all the time is how I noticed ;_)