The file selection is set for each 'Project'

File sets are can be assigned to one of the 5 available backup triggers

All files or 1 user defined selection.

File set can be assigned to one of the available backup triggers

New files are only added if the user selection is All files.

Filter set can be static or dynamically set during the backup session.

[Useful for shared backup cards or sharing a card between multiple devices]

There is only one main backup folder on the SD card and this holds all the 'Project' folders. Each backup folder within a project's folder is name: {Hotsync ID} {Timestamp}

This allows for multiple users to share the same project settings. Each users is allocated the number of backup sets defined in the project settings.

Each hotsync user gets their own folder with in the main BBVFS backup folder on the card. The user's folder holds sub-folders for all their checkpoints, archives and backed up files

Each hotsync user gets their own folder with in the main BBVFS backup folder on the card. The user's folder holds sub-folders for all their checkpoints, archives and backed up files

1. Normal
2. Maximum
3. Fastest

1. Quickly
2. Thoroughly

1. Quickly
2. Thoroughly

[Note: do not use the same password for ZIP encryption as for other stronger encrypting applications, as their effective encryption strength will be exactly that of the ZIP encryption]

Is this sufficient to protect personal data?

Use of password/data protection applications should be considered, such as one using AES, IDEA etc.

Claimed to be weaker than their Non-Zip based compression backup scheme

[Still wise not to use the same password as used for other application: a 56bit key is not as secure as 128 / 1024 / 2048bit]

XTEA ECB with 56bit:claimed to be better than ZIP, but effective strength is unknown.

TEA ECB with 56bit:claimed to be better than ZIP, but effective strength is unknown.

[With multiple schedules / triggers it is still wise to plan to have only one triggering action for a given scenario]

The following are set for each 'Project', so multiple backup schemes can be active and gives a flexible approach to backup strategy: for example, this allow for an incremental, uncompressed snapshot triggered by exiting important apps, incremental, compressed daily backups for the previous week, and a full, compressed backup every week. Giving 9 'Backup Sets' with known data configurations.

There is only one backup scheme active at any one time, that can be activated by many triggers. Backups are normally incremental, with full backups being performed only after a 'Checkpoint' event. 'Checkpoint' events can be automatically triggered, but they may happen at unwanted times.

There is only one backup scheme active at any one time, that can be activated by many triggers. Backups are normally incremental, with full backups being performed only after a 'Checkpoint' event. 'Checkpoint' events are manually triggered,

A schedule is a 'Project' and each defines the files on which it works and the triggers that kick-off the backup: as such, 'Triggers' are local to each 'Project'. Files are backed up to the / PALM / Backups / {Project Name} / {Backup Set folder}

Each trigger is global but can be assigned to backup a different user defined 'Set' of files to the common / Palm / Backup / {Hotsync ID} / Backup folder

Each trigger is global but can be assigned to backup a different user defined 'Set' of files to the common / Palm / Backup / {Hotsync ID} / Backup folder

Just select all days in the above feature

Activated when the global trigger is enabled or if a power-off is performed when a 'Continuous Backup' application is active

[It is best to limit this to a select few application as it can quickly become very annoying]

Globally set with 'Continuous Backup'. This sets the files which activate this trigger

The wait time is set between 1 minte and 99 days 23 hours and 59 minutes. The trigger repeatedly actives between HotSyncs.

The wait time is set between 1 minte and 99 days 23 hours and 59 minutes. The trigger repeatedly actives between HotSyncs.

A repeating period can be set between 1 and 12 hours

A 'Project' is set to be 'Update' backup

Backups are incremental except when a defined 'Checkpoint' threshold is actived

All backups are incremental

A 'Project' is set to be 'Full' backup

A full backup is performed after a 'Checkpoint' is saved

Only full on first use.

Each 'Project' can be set to save an unlimited or a defined number of 'Backup Sets'. The backup sets are cycled, if a defined limit of sets is specified, so that, when the maximum number of sets have been created, the oldest set is used for the next backup session. These are either incrementally maintained 'Update' sets that build on the older backup session files or a completely new 'Full' backup is performed after deleting the old backup's saved files.

With incremental backups, the impact is to minimise the disruption to the user while maintaining multiple, historical versions of the the selected files from the RAM: a 'Full' backup can be very time consuming.

'Checkpoint's of the current backed up files can be manually created or automatically triggered to happen every N days and/or N backups. The number of 'Checkpoint's can be limited by the user if required. The space used by 'Checkpoint' folders can be limited to a set value of MB.

During a 'Checkpoint' event it seems that the backup folder is renamed with a timestamp and then a new empty backup folder is created. This forces that full backup is performed after the 'Checkpoint' and depending on the trigger (such as N backups) this could be at an inconvenient time.

'Checkpoint's of the current backed up files are manually created.

During a 'Checkpoint' event it seems that the backup folder is renamed with a timestamp and then a new empty backup folder is created. This forces that full backup is performed after the 'Checkpoint'.

Tree-viewing allows the selection of the application which selects all of that creator ID's files.

Tree-viewing allows the selection of the application which selects all of that creator ID's files.

Tree-viewing allows the selection of the application which selects all of that creator ID's files.

Tries to use the saved password.

Tries to use the saved password.

The better solution would be to allow the user to decide if the stored password is to be tried during restore: this would have to be an option on the dialog screen used to type in the password, and not a generally accessible option.

Defaults for a full restore.

The previously selected files are still selected. For a partial restore, this could be an issue if they are not visible in the list window. It is advisable to always do a 'Select None' first and then select the files to restore.

The previously selected files are still selected. For a partial restore, this could be an issue if they are not visible in the list window. It is advisable to always do a 'Select None' first and then select the files to restore.

There is no 'basic' mode that allows new users to bypass the more advanced options and concepts of backing up.

There are New Backup and Restore…buttons, but also an Update button. Plus others for Verify, Setup, Sets, and Schedule.

A basic mode should allow a new user to instantly understand what to do, and so make a first backup 'now'. Then there is time to learn about other features.

The most important thing is to do backups [and be able to restore from them, too]; an easy entry to do a first backup is essential so as not to frighten away users.

Basic and Advanced user modes.

The Basic mode is equivalent to the 'lite' version of BBVFS. This gets any user backing up straight away with the big Backup button - almost impossible to go wrong. There is a button for accessing Preferences [schedules, and compression and encryption] and another for a full restore.

Advanced mode provides access to archiving, checkpoints, backup sets, partial backup/restore.

Basic and Advanced user modes.

The Basic mode is equivalent to the 'lite' version of BBVFS. This gets any user backing up straight away with the big Backup button - almost impossible to go wrong. There is a button for accessing Preferences [schedules, and compression and encryption] and another for a full restore.

Advanced mode provides access to archiving, checkpoints, backup sets, partial backup/restore.

Caters for newer Palms with options to set excluded files, and address various PIM and NVRAM specific issues related to backing up: DbCache, newer PIM DBs, locking risky apps.

These options are useful but time is needed to understand which are most appropriate for the user's Palm.

Caters for newer Palms with excluded files and locking risky apps.

This is overtly more appropriate to newer devices than V3, whether this is truly the case or not, it would be better to use V4 for peace of mind.

No specific options to cater for newer Palms: though many updates have been issued since its release, it is possible that many issues are addressed without need to have user options.

No locking of risky apps nor an excluded files list.

Upgrading V3: $11.95

The Resco Backup timings are based on the time reported by the application [I checked and verified that they reported the times as stamped in the backup folder and its files]. For BBVFS I used the first backed up file's creation data and the last backed up file's modification time and calculated the elapse time: this omits the time for checking for orphan files, which is an unnecessary activity in full backups.

Note: Tests are specific to the composition of files resitent in RAM, the Palm device, and the SD card used. Tests on other environments will yield different results.

VFSMark Results for ByteStor 512KB SD Card used in the tests:

• File Create: 50%
• File Delete: 29%
• File Write: 6%
• File Read: 760%
• File Seek: 1475%
• DB Export: 29%
• DB Import: 953%
• Record Access: 905%
• Resource Access: 853%

      VFSMark: 562

VFSMark Results for SanDisk Ultra II Plus 1GB SD Card used in the tests:

• File Create: 163%
• File Delete: 141%
• File Write: 109%
• File Read: 760%
• File Seek: 1311%
• DB Export: 54%
• DB Import: 811%
• Record Access: 905%
• Resource Access: 853%

      VFSMark: 567

Tests where also performed without any compression at all, and for both applications these tests were significantly slower than the results with compression. As such I will be backing up with compression switched on.