AdamS

Phil what software are you using to open the PST file after you have created it? I couldn't duplicate precisely what you have shown there but I did have the _files appended to the root archive only, by that I mean the OST file I used as a test had the _files appended to the end in the new PST I created from a selection of the emails. Otherwise all other folders (inbox, outbox, sent etc) appeared normal. The pic below is as previewed in Outlook but even using Kernal PST viewer still shows the same.

You have been busy! Looks fantastic, can't wait to get my hands on this and see how it all works. Will definitely have some thoughts once I've had a chance to use/test but at the moment can't think of anything. Edit: any clues on how far away the release is?

The method Alex describes is exactly what I do on my larger Connect cases, you can also run the keyword searches required then highlight and tag off the batches yourself using parent/child tags. The flag column lists numbers starting at 1 so it's easy enough to scroll down to 1000, highlight, then create your tags (Parent --> batch 1, Parent-->batch 2 etc). I'm running a review case at the moment where I tagged up a few hundred thousand files using 16 different keywords in batches of 1000 for my client, took me less than 30 mins to pre batch it like that for them.

Is there any issues with bulk untagging? I'm finding that selecting large amounts (313k or 175k) of items and untagging all seems to stall Intella. It doesn't crash, just sits there with the little progress window moving along but nothing actually happening that I can tell. I close the case and then reopen and all items are still tagged.

Hi Keaton, I've had some similar thoughts myself and this is the approach I've taken. Exception items I will note the files and depending on the size/number I'll either go to the original source documents and then copy them into their own folder called 'Exception Reprocessing', or export them directly from Intella. Then I will attempt to correct the issues that stopped Intella processing them in the first place (OST's are quite common here so ScanPST can be helpful). I'll generally set up a 'test' intella case seperate from the main case to test the exception items until I'm satisfied I can process them all, or as many as possible. I'll then add the 'Exception Reprocessing' folder as a new source and index it merging it with the main case. With regards to the PDF docs and OCR, Intella has no native OCR capability, the way to locate PDF's which cannot be searched is with the 'empty documents' flag, you can see that directly under the 'features' facet, from there you can highlight the documents you want to export for OCR, OCR them with your tool of choice, they can then be imported back into Intella and they will be searchable.

I'm working on a few cases at the moment where I'm having to do a lot of 'pre review' tagging for the client. To facilitate this I'm doing a lot of tags to build search queries then I go back and remove tags from certain items and later delete other tags etc. No problems there, but I have noticed that the parent/child relationship seems to throw out some of the tag menus when it comes to showing the data alphabetically. The tag menu looks perfect, but if I highlight all my tags and click search they are displayed randomly in the search box and the parent/child relationship is not visible. My tags are all a particular keyword as the parent, then the children are either 'email' or 'document' (in this case) but they can be anything. The search window shows them all randomly with no clear indicator which parent each of the 'email' and 'document' tags belong to. Likewise when it comes time to remove tags from some items, if I highlight a bunch of items then go to the 'remove tags' menu from right click the order is haphazard (but the parent child relationship is visible) with children appearing randomly nowhere near their parents. Every one knows children should always stay close to their parents for safety So, long wish is to have the Tag menu keep it's alphabetic display and parent/child relationship just as it's shown in the tag facet menu.

The auto tag keyword process is the best thing since sliced bread....well it could be If the ability to apply some filters to that process was added then it definitely would be, I'm thinking specifically dedupe and one other facet filter,the same way we can when viewing. Add before tag then away you go.

I've matched MD5 from Xways to Intella with no issues, as far as I understand the mathematical concept MD5 produced by any tool will work as they are applying the same algorithm to the same data...at least it should.

Phil as it happens I've spent the last few days delving very deeply into this in order to recover a case that went belly up after a Windows BSOD. You can indeed import an MD5 hash list, it's one of the facet options right near where you can import keyword lists, and it works in the same fashion, you can import a .csv file with one MD5 per line and must have a heading MD5 hash (if you want to see the format just export a CSV with Intella and only select the MD5 value. To the hash question, I would wait for confirmation from the admins but my suspicion is the MD5 is the entire file (email including attachments) and the message hash is either just the body of the email, or the email sans attachments. I always use the MD5 hash when importing/exporting file lists.

You're right it does, I'd forgotten about that post. I did try restoring annotations with the events.log file but was unsuccessful, however I didn't convert the file to CSV, I must have missed that in the manual. In this case a reindex of the data appeared to fix the issue, although I did lose 2 days worth of tagging due to my bad backup software issues.

I'm not sure exactly what to call this but I've had a few experiences where if the machine crashes then Intella seems to take it very badly, particularly if it's in the middle of tagging or indexing. Indexing is not such a major issue because you can simply start that process again and nothing is lost but time, however if a case becomes corrupt whilst tagging there is the potential to lose days/weeks of work. The crashes I'm referring to are nothing to do with Intella, it's usual windows BSOD stuff, but Intella is the innocent victim of this drive by. Is there something that can be done to make Intella more robust when dealing with system crashes and the like? Putting aside the fact that this can be avoided by us using a proper backup system (which I do) but as has happened to me in this case my backup software turned out to be woefully inadequate and best case scenario I may have lost a day's worth of my clients work...worst case nearly 2 weeks!!

I'm having an issue with a windows crash corrupting my case ( I have emailed support regarding this) and something occurred that may assist. Is there a log file somewhere which details all the user activity? My thinking here is that if I have to go back to an earlier restore point from my backups I'm going to lose some of the client's work. If there is a log of all the items tagged then I can reproduce the lost work for the client once the older version of the case is up and running.

Thanks Christian, these are mostly client driven feedback so there will always be a certain amount of "trying to make it more like another tool" but all in all the responses has been very positive. Currently we can only PIN 3 of the quick slots, it would make sense to lift that number to 9 if possible.

Perfect thank Lukasz, now worries on the updated build that is a minor issue that can wait. I'll let my client know about pinning the tags thought that's what they really want.

Okay, sorry for the new post but the other one was getting a bit messy. Clearing the cookies reset the connect view back to default. Is there any documentation or could you provide me with a basic explanation of how the quick tag behaves in Connect. I have been playing around with it and it's clear there is a pattern to the items that are displayed but it alludes me. At first I thought it was dynamic but it seems to not update under some circumstances and does on others. What my client was experiencing was that a particular tag would drop off the list when another tag was used, however if they used the tag that had dropped off it was not making it's way back onto the quick tag list no matter how often they used it. They have one particular tag that they need to use most often and as they are reviewing thousands of documents it's quite pains taking to apply the tag manually in each instance.

I found this today. I was testing something else and created 11 test tags, then tagged some random emails to see how the quick tag list behaved as my clients are having a problem with a tag dissappearing from the quick tag list once it's used. I couldn't duplicate their problem so I'll address that in a support email. But, what I did notice was that when I'd finished testing and deleted all the test tags I made, they still appear in the quick tag list when I open a preview window for an email. Log out and in doesn't clear them and disable Connect server and restart doesn't clear them either. They appear to be stuck on screen. Can this manually be cleared somewhere? Edit : disregard, closing the browser and restarting cleared the quick tag field, obviously stored in temp cache somewhere. Eit again : actually no it didn't...not sure what happened there but browser close down and clear cache doesn't clear those settings.

ahhhhhhh Yes that would make sense. I'm starting to get more and more requests for this type of 'locked down' tag selection so that would work perfectly. My current job the lawyers have 15 tags they are using and having a preset tag list like this visible would be great, they could select a field and move on to the next one.

I agree on the locking duplicates button and have already explained to them that there is no way to determine which duplicate(s) will be shown/hidden. Some more requests This sort of goes in hand with default settings but the quick tag behaviour seems to be dynamic in that it changes to reflect the most recently used tags rather than being static. I think I'd previously suggested increasing the number of quick tags beyond 9, the numpad could be used for the first 9 then it would have to be mouse click for the others, but increasing the number would enable the tags to remain static without losing any functionality. Currently my clients are reviewing with 12 set tags but they have 1 which is vital and it keeps vanishing off the quick tag list when other tags are used. In the short term is there a way to assign a tag to stay in the quick tag view? Retain keyword information - If I run a keyword list or search in the initial results the applicable keyword is highlighted in certain views, however if I apply a tag to the responsive emails, clear the search results then view the data via the tag the keyword information is lost. This type of information is not always going to be relevant so I don't think this should be a default behaviour, but perhaps a 'retain keyword hits' radial button when applying tags. As I think on it I realise this would be difficult as not all searches will result from keywords, but some way to retain this data would be beneficial. Have tag information visible when 'raw data' tab is selected. In some cases raw data is the only view available and with this view there is no information to show existing tags. The only way appears to be by selecting 'tags' and seeing if any ticks are in boxes. Perhaps make this information permanently visible on the left hand side with its own little banner field below the print/save etc buttons? When applying tags the normal tag field has predictive capability, but the parent tag is a static drop down list. Can get difficult to navigate when you have hundreds of tags to choose from, dual capability here would be fantastic. Drop down list with text field at the top that has predictive capability so we can just start typing the parent tag we want and have the drop down list dynamically change according to what we type.

Any feedback on these gents? My client keeps asking..

Sounds good Lukasz, I have a few Connect cases running at the moment and in general the feedback is very positive (especially compared to the 'big player') so these type of customisations are helping seperate and elevate (not a bra advert) Connect from the other players. On that note, some more user requests. 1 - The ability to customize some of the default parameters. Specifically the request was the ability to lock the 'deduplicate' button so that it could not be 'un-deduplicated'. They wanted some way to assure themselves that once set it could not be inadvertantly unset. More a piece of mind thing than anything, but it got me thinking about the ability to customize some of the default settings and behaviour of Connect when we set up the case. Some sort of config options or config file that we (the administrator) could set up for our client to customize the view and some options. For me the customization is a very strong point, but some of my clients are stuck in the ways of other tools and want to be able to duplicate that behaviour (to a fashion) to remove the possibility of accidentally changing things. 2 - This could also be taken as a question. Can we reset the stats? My clients are about to start a review on a historical job and it would be handy if we could remove the 'preview' status on all items so they can start fresh and track which items they have previewed. Presently many of the items have been previewed by me on the original analysis.

hmmm templates you say.....probably past time I looked in to how they work

Sorry I don't understand the second request?

Disregard I figured it out. I had to reindex with the latest version before I could OCR.

I'm viewing an old case that was created in 1.6.x (I think, about 2 years ago) with 1.8.4 and I want to export and OCR some PDF docs, but the export for OCR option is greyed out. Do I need to reindex for this to be available in the newer version?

Something I've wanted for a very long time but because it's purely cosmetic I've not really pushed it, but have a job at the moment which requires a lot of small tags be applied so it's really highlighted for me the extra time that is required to view data under some circumstances. Would it be possible to have a 'double click' mimic pressing the search button for any of the facets? (not the main search button for keywords). At the moment if I want to view any given tag I have to highlight the tag then click search, then click on the ball that appears to populate the preview field. I know it sounds really pathetic and lazy when I type it out like that, but trust me when you are applying and editing hundreds of tags for small batches it becomes a long process. Double click on a tag and BOOM the ball and preview fields instantly populate, you'll thank me later