What Does Transparency Mean?

One of the things I love about making art is that it often inspires me to make connections I had not made before. I’m currently working my way through Julie Balzer‘s A Year of Gelatin Printing class, and there’s a month called “Layer Layer Layer” that includes a lesson on transparency.

The first technique was simple, as the outcome was not meant to be the artwork. In order to assess whether or not a particular tube of paint was transparent, the lesson was to grab a book page* and apply paint to it, to see whether or not the paint is opaque, semi-transparent, or transparent.

It is not a particularly challenging lesson, which gave me time to think about when I might want different levels of transparency, what kinds of layering I might do.

a book page. The middle has no paint, the top has a layer of transparent orange, the bottom has a layer of semi-transparent red.
Transparent at the top, semi-transparent at the bottom

Suddenly my brain leapt to a realization – transparency in a business context is often stated as a binary. I’m a huge fan of smashing binaries to get the rich matrix that lies beneath, and this was no different.

Transparency can mean:

  • Decisions, and reasons for making those decisions, are talked about openly.
  • Being open and honest about bad news, a project going off track, or a timeline change, instead of covering these up.
  • Recognizing those who supported your effort, helped with research, or lent their eyes and opinions to make your work better.
  • Sharing your roadmap publicly, and sharing any changes publicly
  • Agreeing with a coworker or customer who has complaints about your product
  • Being forthcoming about product or team limitations

Of course, transparency can be taken too far. There are plenty of reasons for privacy – discussions with HR (including hiring, retention and losing employees), not telling employees about a big change (e.g. merger) until you know it’s going to happen so uncertainly doesn’t spread, and of course personal details such as government ID numbers, bank information for direct deposits, and authentication credentials.

Transparency is not simply a matter of making information available, though. Good transparency is not just a permissive shared folder or wiki; good transparency ensures that people are well-informed.

For example: let’s say the 50-page employee handbook changes. Transparency is on a spectrum, with one end being an oversimplified notification:

Please note, the employee handbook has changed for US employees.

This is the equivalent of making everything public and letting people sort out for themselves not only which information is significant, but what, if any, changes there have been to call awareness to. It’s not particularly useful for transparency – it’s opaque, like the top section of black paint on this book page:

a book page, with an opaque thick black strip at the top, the middle has no paint, and the bottom is semi-transparent through brown paint
Opaque at the top, semi-transparent at the bottom

Semi-transparency, such as the brown paint at the bottom of the page in the image above, might be to explain what has changed:

Please note, the employee handbook for US employees has been changed to reflect the following policy changes:

– “Hairstyle” and “hair texture” are now explicitly named as protected categories under discrimination, harrassment, and equal opportunity employment.

– There is a new section explaining the ban on firearms at the workplace.

Also known as, “Please don’t make me read 50 pages to figure out what changed!”

I pointed out that this might be semi-transparent, as it depends on how it’s done, and if it accurately portrays important changes and downplays unimportant changes. Of course, what is important can be different for different departments, teams and people at an organization, so this can be a bit trickly.

Similarly semi-transparent is highlight what changed – this method calls out exactly what changed, like a version history or file diff. If there are very few changes, this can illuminate what changed better than a description of the changes. However, if there are a lot of changes, a summary might be better.

What would the most useful form of transparency be? Well, it depends on the recipient as well. I would prefer a mixture of the three, if there are a lot of changes; if there are a few significant changes, highlighting would suffice; and if there are many repetitive changes, an announcement is fine – e.g. “all instances of dog have been changed to pet”.

I’m sure we all have stories of how too much transparency caused issues – not being able to find a document or conversation because all docs and conversations are public.

* I have an old copy of Head First HTML 5 Programming I use for painting on book pages

MariaDB Foundation vs. Open Database Alliance

So lots of folks are talking about the newly announced MariaDB Foundation. Some folks are confused as to what happened to the Open Database Alliance?

The MariaDB Foundation members include Monty Program, SkySQL, David Axmark, and Allan Larsson.

The Open Database Alliance members include Monty Program, FromDual, Open query, Percona and SoftMethod. [SkySQL was not founded when the ODBA was announced.]

The mission of the MariaDB Foundationwell, its not on the MariaDB Foundation web page yet, but the blog announcement says:

In its mission statement, the MariaDB Foundation exists to improve database technology, including standards implementation, interoperability with other databases, and building bridges to other types of database such as transactional and NoSQL. To deliver this the Foundation provides technical work in reviewing, merging, testing, and releasing the MariaDB product suite. The Foundation also provides infrastructure for the MariaDB project and the user and developer communities.

The mission of the Open Database Alliance (from http://odba.org/about/):

The Open Database Alliance (ODBA) is a non-profit organization dedicated to the promotion of the commercial and non-commercial ecosystem around Open Source Database Technologies. Whereas proprietary software forces strong dependencies upon a single vendor, Open Source (a.k.a. Free Software, Libre Software, FOSS or FLOSS) enables an ecosystem of commercial offerings around certain technologies, e.g. GNU, Samba, Apache, the Linux Kernel.

Databases are an essential base technology for most higher applications, from desktop search over web services to the entire spectrum of enterprise IT. Enabling, growing and protecting that ecosystem for all Open Source Database Technologies are primary activities of the ODBA.

According to h-online, there are already 1 million euros pledged to the MariaDB Foundation for this.

According to the ODBA fee schedule, a Silver member pays at least US$7k (depending on how many employees the company has) and a Gold member pays at least US$35k per year.

There is nothing on what the Open Database Alliance has done, or is currently doing, or is planning to do, other than doing the work of a business analyst and getting referral fees for it. And certainly, theres nothing concrete.

The MariaDB Foundation has nothing concrete planned, but theyre fairly new. And their mission makes it clear that their organization focuses on MariaDB, whereas the ODBA is vague, and could apply to Perconas patched MySQL, or Oracles version of MySQL.

If the MariaDB Foundation succeeds in getting 501(c)3 designation, they will have to be extremely careful they do not lose it. There are huge potentials for conflicts of interest (for example, does a bounty get paid by the MariaDB Foundation or the Monty Program?), which is grounds for revoking a 501(c)3 designation. (I know, because in the past I have been the Treasurer for national and international organizations, and currently server as the Treasurer for Technocation and a local fiber guild.)

With such similar members, what is the MariaDB foundation doing that the Open Database Alliance cannot do especially considering http://odba.org/legal/ says MariaDB is a trademark of Monty Program Ab. The Open Database Alliance and its members has broad usage rights of this trademark?

What happened to all that money that went into the Open Database Alliance? What is the MariaDB Foundation going to do with all the money it is gathering?

Is the Open Database Alliance even a valid organization any more? It was founded May 13, 2009 well over 3 years ago. What happened to it?

Backups Video Now Downloadable

The Google Video for the MySQL Backups presentation stops after about 13 minutes, so I’ve put it up as a download at:

http://www.technocation.org/videos/BostonMySQLMeetupJanuary-mp4.mov

Many of the questions brought up by the discussion of the summary slide are answered in the presentation.

As always, I encourage feedback — both technical as well as presentational (ie, the slides were not understandable, you talk to fast, etc).

The slides are up at:

http://sheeri.net/presentations/MySQLbackups.swf — 125 Kb flash
or
http://sheeri.net/presentations/MySQLbackups.pdf — 2 Mb PDF

Why didn’t I think of this? Tom Limoncelli was nice enough to create a livejournal feed for this site — http://www.livejournal.com/users/sheericom_feed/

Direct link to add as a ‘friend’ if you’re logged in — http://www.livejournal.com/friends/add.bml?user=sheericom_feed

Why didn’t I think of this? Tom Limoncelli was nice enough to create a livejournal feed for this site — http://www.livejournal.com/users/sheericom_feed/

Direct link to add as a ‘friend’ if you’re logged in — http://www.livejournal.com/friends/add.bml?user=sheericom_feed

The MySQL User Group meeting on Monday, January 9th will focus on backups — I will give a 30-45 minute presentation on backups and we will then delve into a discussion — how folks within the group are backing up, how they’re not backing up, how they’d like to back up. If there’s time, we’ll do a “lessons learned from backup horror stories”.

Afterwards, we’ll head a few doors down to Boston Beer Works for more chatting.

You can see a video of last month’s presentation by Philip Antionades of MySQL AB on the new features in MySQL 5.0 at Google Video:

http://video.google.com/videosearch?q=Mysql+5.0+by+Philip+Antoniades

(we will video this month’s presentation but not the discussion)

More details, including the event location and directions, are at: http://mysql.meetup.com/137/events/

Any questions, comments, etc should be sent to me. We’re currently looking for a space better than North Station — something around MIT might be nice, and we’re trying to get an MIT grad student/faculty/staff member/organization to sponsor it. (We want a location that’s good for people driving AND taking the T, and when the Garden has an event, it’s not feasible to park there). If you can help out, please let me know.

(You may forward this announcement to other groups, lists, blogs, whatever.)

I’m quite excited about the presentation, as I’m writing down a lot of information from different sources that really wants to be together, but as far as I can tell is not yet, at least not in a public place.
Why didn’t I think of this? Tom Limoncelli was nice enough to create a livejournal feed for this site — http://www.livejournal.com/users/sheericom_feed/

Direct link to add as a ‘friend’ if you’re logged in — http://www.livejournal.com/friends/add.bml?user=sheericom_feed

The MySQL User Group meeting on Monday, January 9th will focus on backups — I will give a 30-45 minute presentation on backups and we will then delve into a discussion — how folks within the group are backing up, how they’re not backing up, how they’d like to back up. If there’s time, we’ll do a “lessons learned from backup horror stories”.

Afterwards, we’ll head a few doors down to Boston Beer Works for more chatting.

You can see a video of last month’s presentation by Philip Antionades of MySQL AB on the new features in MySQL 5.0 at Google Video:

http://video.google.com/videosearch?q=Mysql+5.0+by+Philip+Antoniades

(we will video this month’s presentation but not the discussion)

More details, including the event location and directions, are at: http://mysql.meetup.com/137/events/

Any questions, comments, etc should be sent to me. We’re currently looking for a space better than North Station — something around MIT might be nice, and we’re trying to get an MIT grad student/faculty/staff member/organization to sponsor it. (We want a location that’s good for people driving AND taking the T, and when the Garden has an event, it’s not feasible to park there). If you can help out, please let me know.

(You may forward this announcement to other groups, lists, blogs, whatever.)

I’m quite excited about the presentation, as I’m writing down a lot of information from different sources that really wants to be together, but as far as I can tell is not yet, at least not in a public place.
This article is somewhat long. Interestingly, as there is much to talk about besides the mechanics of each backup option. I wonder what I’d need to do to make this into a white paper or an article?

The backup presentation was finished last night. I may decide to go back and put some extra stuff in there, but that would be syntax and code and stuff. The logic is all in there, and the notes have been printed. I will post the slides (in .pdf and .swf (flash, the file is very small that way) formats) after the talk on Monday, as I may yet revise them.

I am very excited about one slide in particular, and I’ll share it here. It’s really a slide that I end with, but I feel as though it’s a great starting point as well as a summary point. I haven’t seen this information encapsulated this way before, so here goes:

Comparison Table of MySQL Backup Methods

Continue reading “Backups Video Now Downloadable”

MySQL Backup Presentation

The Boston MySQL User Group was successful yet again! I was a bit worried about my presentation, that it would be too basic or folks would have wanted to see actual code and scripts, but it turned out well.

You may notice that the links at the side (http://www.sheeri.net if you’re reading a feed) include a category called “Presentations”. Currently the December presentation is linked to, direct to Google Video, and the slides from tonight’s meeting are also up in PDF format as well as macromedia flash. And, of course, tonight’s video, thanx to Mike Kruckenberg. Mike also took pictures!

Folks who were at the presentation — feel free to let me know what you thought of it. Folks not at the presentation — feel free to watch the video and let me know what you think. I’m especially interested for those folks who read the last article, if the talk goes more in depth and answers more questions.

(quick links for those reading feeds:

Mysql User Group Photos

Mysql Backups (slides, flash, 125 Kb)
Mysql Backups (slides, pdf, 2.0 Mb)

MySQL Backups presentation at the Jan. 9th, 2006 meeting.
Mysql 5.0 Presentation at the December user group meeting featuring Philip Antoniades of MySQL AB.
MySQL webinars from mysql.com
)

Why didn’t I think of this? Tom Limoncelli was nice enough to create a livejournal feed for this site — http://www.livejournal.com/users/sheericom_feed/

Direct link to add as a ‘friend’ if you’re logged in — http://www.livejournal.com/friends/add.bml?user=sheericom_feed

Why didn’t I think of this? Tom Limoncelli was nice enough to create a livejournal feed for this site — http://www.livejournal.com/users/sheericom_feed/

Direct link to add as a ‘friend’ if you’re logged in — http://www.livejournal.com/friends/add.bml?user=sheericom_feed

The MySQL User Group meeting on Monday, January 9th will focus on backups — I will give a 30-45 minute presentation on backups and we will then delve into a discussion — how folks within the group are backing up, how they’re not backing up, how they’d like to back up. If there’s time, we’ll do a “lessons learned from backup horror stories”.

Afterwards, we’ll head a few doors down to Boston Beer Works for more chatting.

You can see a video of last month’s presentation by Philip Antionades of MySQL AB on the new features in MySQL 5.0 at Google Video:

http://video.google.com/videosearch?q=Mysql+5.0+by+Philip+Antoniades

(we will video this month’s presentation but not the discussion)

More details, including the event location and directions, are at: http://mysql.meetup.com/137/events/

Any questions, comments, etc should be sent to me. We’re currently looking for a space better than North Station — something around MIT might be nice, and we’re trying to get an MIT grad student/faculty/staff member/organization to sponsor it. (We want a location that’s good for people driving AND taking the T, and when the Garden has an event, it’s not feasible to park there). If you can help out, please let me know.

(You may forward this announcement to other groups, lists, blogs, whatever.)

I’m quite excited about the presentation, as I’m writing down a lot of information from different sources that really wants to be together, but as far as I can tell is not yet, at least not in a public place.
Why didn’t I think of this? Tom Limoncelli was nice enough to create a livejournal feed for this site — http://www.livejournal.com/users/sheericom_feed/

Direct link to add as a ‘friend’ if you’re logged in — http://www.livejournal.com/friends/add.bml?user=sheericom_feed

The MySQL User Group meeting on Monday, January 9th will focus on backups — I will give a 30-45 minute presentation on backups and we will then delve into a discussion — how folks within the group are backing up, how they’re not backing up, how they’d like to back up. If there’s time, we’ll do a “lessons learned from backup horror stories”.

Afterwards, we’ll head a few doors down to Boston Beer Works for more chatting.

You can see a video of last month’s presentation by Philip Antionades of MySQL AB on the new features in MySQL 5.0 at Google Video:

http://video.google.com/videosearch?q=Mysql+5.0+by+Philip+Antoniades

(we will video this month’s presentation but not the discussion)

More details, including the event location and directions, are at: http://mysql.meetup.com/137/events/

Any questions, comments, etc should be sent to me. We’re currently looking for a space better than North Station — something around MIT might be nice, and we’re trying to get an MIT grad student/faculty/staff member/organization to sponsor it. (We want a location that’s good for people driving AND taking the T, and when the Garden has an event, it’s not feasible to park there). If you can help out, please let me know.

(You may forward this announcement to other groups, lists, blogs, whatever.)

I’m quite excited about the presentation, as I’m writing down a lot of information from different sources that really wants to be together, but as far as I can tell is not yet, at least not in a public place.
This article is somewhat long. Interestingly, as there is much to talk about besides the mechanics of each backup option. I wonder what I’d need to do to make this into a white paper or an article?

The backup presentation was finished last night. I may decide to go back and put some extra stuff in there, but that would be syntax and code and stuff. The logic is all in there, and the notes have been printed. I will post the slides (in .pdf and .swf (flash, the file is very small that way) formats) after the talk on Monday, as I may yet revise them.

I am very excited about one slide in particular, and I’ll share it here. It’s really a slide that I end with, but I feel as though it’s a great starting point as well as a summary point. I haven’t seen this information encapsulated this way before, so here goes:

Comparison Table of MySQL Backup Methods

Continue reading “MySQL Backup Presentation”

Backups

This article is somewhat long. Interestingly, it does not actually cover my entire talk, as there is much to talk about besides the mechanics of each backup option. I wonder what I’d need to do to make this into a white paper or an article?

The backup presentation was finished last night. I may decide to go back and put some extra stuff in there, but that would be syntax and code and stuff. The logic is all in there, and the notes have been printed. I will post the slides (in .pdf and .swf (flash, the file is very small that way) formats) after the talk on Monday, as I may yet revise them.

I am very excited about one slide in particular, and I’ll share it here. It’s really a slide that I end with, but I feel as though it’s a great starting point as well as a summary point. I haven’t seen this information encapsulated this way before, so here goes:

Comparison Table of MySQL Backup Methods

Method No Locking DDL Snapshot Remote Free All Engines All Tables Text File Recover
Corruption
# No
SELECT . . .
INTO OUTFILE
Engine
Dependent
No No Yes Yes Yes No Yes Yes 3-4
mysqldump No Option No Yes Yes Yes Yes Yes Yes 2-3
Replication Yes No Yes Yes Yes Yes Yes No No 3
OS level copy No Yes No No Yes No Yes No Yes 5
mysqlhotcopy Yes Yes Yes No Yes No Yes No Yes 3
InnoDB Hot
Backup
Yes Yes Yes No No No Yes No Yes 4

The table is set up so the “No” answers are ‘bad’ and the “Yes” answers are ‘good’. The last column contains the # of “No”s in the row. Right off the bat, we see why mysqldump is probably the most used backup tool — it has the fewest “No”s.

“No locking” — A backup routine should interfere as little as possible with the actual database. If you lock a row or table, even for just reading, you’re blocking others from writing it. Transactional engines like InnoDB and BDB offer transactions to help get around this, but many databases have a mixture of transactional and non-transactional storage engines.

“DDL” — Backups should be able to completely restore a database. If you do not have the Data Definition Language (ie, CREATE TABLE statements), you will not be able to completely restore a database.

“Snapshot” — To use a backup for point-in-time recovery or to build a new replication slave, a snapshot is needed. It is possible to do a point-in-time recovery without a snapshot, but it involves checking the binary logs for possible duplicate statements.

“Remote” — Does the backup need to be run on the OS where the server lies? Due to security in companies or the use of an ISP, it is not always possible for the DBA to have the level of access needed to run programs on the database server.

“Free” — Self-explanatory. Note that the backup options that are free come packaged with MySQL, with the exception of OS-level copy, which is in all OS systems (ignoring embedded systems for now).

“All Engines” — Does the backup option deal with all storage engines? Some tools or commands are engine-specific.

“All Tables” — Can the tool easily backup all the tables without external looping code?

“Text file” — A text file backup is advantageous for a two reasons: Corruption can be easily detected, partial backups (ie, of one table) can be done, and SQL can be standardized for migration. Text files are usually larger than data files, but compression of text is excellent.

“Recover Corruption” — A “Yes” here means the backup option can be used to recover from corruption.

A brief discussion about the options:

SELECT . . . INTO OUTFILE is not used often, because it does not copy DDL. Code is needed to loop through tables, and there will be table-level locking for MyISAM tables. If the SELECT statement is put into a transaction, there will not be a problem for BDB and InnoDB tables.

In MySQL 5.0 it is possible to copy the information about a table from the INFORMATION_SCHEMA database by using SELECT . . .
INTO OUTFILE
. I have not tested if inserting data into the INFORMATION_SCHEMA tables is an equal and complete substitution for DDL. Instinct says “no”.

It’s my opinion that mysqldump is the most widely used backup tool. It has the fewest “No”s of all of them, and the features it does not have can be worked around.

Note that mysqldump does do DDL by default, but it’s easy to get an incomplete DDL statement. If you turn off –opt, and don’t put –create-options, you will end up with CREATE TABLE statements that are standard SQL. This may sound good, but remember that storage engines are MySQL specific, so your CREATE TABLE statements will end up using the default storage engine, which may not be what you want upon restoration.

mysqldump can do a snapshot if the option to lock all the tables is used. However, it is usually not feasible to lock every database on the server while the backup is running, because a backup can take minutes.

Replication is also widely used, but it requires another instance. And to reduce single points of failure, that means more hardware needs to be bought. It may be difficult to justify hardware simply to do nothing unless a backup is needed.

One of the features of replication is also a detriment. The standard way to set up a replication slave is to take a snapshot of the master, import it to the slave, and then use binary logs from there. Another way is to use LOAD DATA FROM MASTER, but that only works for MyISAM tables. It is possible to alter tables and data on the slave before or during replication, without replication failing. This is a handy feature for having a write-only master with InnoDB tables and a read-only slave with MyISAM tables (for example, with fulltext searching). However, this also means that the DDL backup is not necessarily to be trusted.

That mistrust can be extended to the data for the same reasons. Replication may not be a reliable backup if DML is being run on the slave server. As well, if the master becomes corrupt, it is likely that corruption will spread to the replication server.

The two most widely-used storage engines, MyISAM and InnoDB, are OS independent (explanation in the MySQL manual for MyISAM and InnoDB). This means that they can simply be copied on the OS-level. The biggest gotcha is that in order to get a snapshot, the entire database has to be locked. And unlike mysqldump, there is no option to an OS-level copy that will lock the tables. Using this is bulkier than mysqldump , and it cannot be used for all storage engines as mysqldump can.

Two solutions have been devised to combat those problems — mysqlhotcopy and InnoDB Hot Backup. mysqlhotcopy is for MyISAM tables while InnoDB Hot Backup is for InnoDB tables. The other difference is that the InnoDB Hot Backup is not free. However, InnoDB Hot Backup does offer a free perl script that will use InnoDB Hot Backup to take a snapshot of the InnoDB tables, and then use MySQL commands to get a snapshot of the other engine types.

Most people will choose a combination of backup methods — many use both replication and mysqldump, or replication and mysqlhotcopy and InnoDB Hot Backup. There is no clear-cut answer, and mostly it depends on your environment and what the backup will be used for.

Ideally there would be one tool that performed all of the functions listed in the chart. The chart is not complete, either — just a listing of my opinion of the most common desired features. There are other desired features not listed, for example replication is an “immediate hot backup” whereas all the other tools require manually importing or starting the MySQL server with new data files, etc. I am definitely interested in what folks have to say on the subject.

Why didn’t I think of this? Tom Limoncelli was nice enough to create a livejournal feed for this site — http://www.livejournal.com/users/sheericom_feed/

Direct link to add as a ‘friend’ if you’re logged in — http://www.livejournal.com/friends/add.bml?user=sheericom_feed

Why didn’t I think of this? Tom Limoncelli was nice enough to create a livejournal feed for this site — http://www.livejournal.com/users/sheericom_feed/

Direct link to add as a ‘friend’ if you’re logged in — http://www.livejournal.com/friends/add.bml?user=sheericom_feed

The MySQL User Group meeting on Monday, January 9th will focus on backups — I will give a 30-45 minute presentation on backups and we will then delve into a discussion — how folks within the group are backing up, how they’re not backing up, how they’d like to back up. If there’s time, we’ll do a “lessons learned from backup horror stories”.

Afterwards, we’ll head a few doors down to Boston Beer Works for more chatting.

You can see a video of last month’s presentation by Philip Antionades of MySQL AB on the new features in MySQL 5.0 at Google Video:

http://video.google.com/videosearch?q=Mysql+5.0+by+Philip+Antoniades

(we will video this month’s presentation but not the discussion)

More details, including the event location and directions, are at: http://mysql.meetup.com/137/events/

Any questions, comments, etc should be sent to me. We’re currently looking for a space better than North Station — something around MIT might be nice, and we’re trying to get an MIT grad student/faculty/staff member/organization to sponsor it. (We want a location that’s good for people driving AND taking the T, and when the Garden has an event, it’s not feasible to park there). If you can help out, please let me know.

(You may forward this announcement to other groups, lists, blogs, whatever.)

I’m quite excited about the presentation, as I’m writing down a lot of information from different sources that really wants to be together, but as far as I can tell is not yet, at least not in a public place.
Why didn’t I think of this? Tom Limoncelli was nice enough to create a livejournal feed for this site — http://www.livejournal.com/users/sheericom_feed/

Direct link to add as a ‘friend’ if you’re logged in — http://www.livejournal.com/friends/add.bml?user=sheericom_feed

The MySQL User Group meeting on Monday, January 9th will focus on backups — I will give a 30-45 minute presentation on backups and we will then delve into a discussion — how folks within the group are backing up, how they’re not backing up, how they’d like to back up. If there’s time, we’ll do a “lessons learned from backup horror stories”.

Afterwards, we’ll head a few doors down to Boston Beer Works for more chatting.

You can see a video of last month’s presentation by Philip Antionades of MySQL AB on the new features in MySQL 5.0 at Google Video:

http://video.google.com/videosearch?q=Mysql+5.0+by+Philip+Antoniades

(we will video this month’s presentation but not the discussion)

More details, including the event location and directions, are at: http://mysql.meetup.com/137/events/

Any questions, comments, etc should be sent to me. We’re currently looking for a space better than North Station — something around MIT might be nice, and we’re trying to get an MIT grad student/faculty/staff member/organization to sponsor it. (We want a location that’s good for people driving AND taking the T, and when the Garden has an event, it’s not feasible to park there). If you can help out, please let me know.

(You may forward this announcement to other groups, lists, blogs, whatever.)

I’m quite excited about the presentation, as I’m writing down a lot of information from different sources that really wants to be together, but as far as I can tell is not yet, at least not in a public place.
This article is somewhat long. Interestingly, as there is much to talk about besides the mechanics of each backup option. I wonder what I’d need to do to make this into a white paper or an article?

The backup presentation was finished last night. I may decide to go back and put some extra stuff in there, but that would be syntax and code and stuff. The logic is all in there, and the notes have been printed. I will post the slides (in .pdf and .swf (flash, the file is very small that way) formats) after the talk on Monday, as I may yet revise them.

I am very excited about one slide in particular, and I’ll share it here. It’s really a slide that I end with, but I feel as though it’s a great starting point as well as a summary point. I haven’t seen this information encapsulated this way before, so here goes:

Comparison Table of MySQL Backup Methods

Continue reading “Backups”