ŁukaszBachman

Hi Adam! Can you elaborate on that part?

Hi Adam! Thanks for sharing those ideas. We had those on our radar but haven't managed to fit it into our Intella 1.9 plans. However we will fix it in 1.9.1.

Would you see that in Desktop, Connect or both? In Connect I could imagine this feature to be more useful and probably would be quicker to develop. What do you think?

Upcoming 1.9 release of Intella and Intella Connect will allow users to sort on a newly introduced "Family Date" column. This will allow to sort all document families (ex. email + attachments) in a chronological order.

@Adam, we are in late stages of development (feature freeze) and testing, so it shouldn't take very long. @Jason, basically there will be two ways of working with data, both based on the premise that Processors and Connect will have access to the same shared resource (disk). Option 1: use drive mapping built into Windows This essentially allows for Connect and Processor to see the same folder mapped as a drive. For instance, you can map some folder so that it appears as a "Z" drive. If this is done consistently on Processor and Connect then "Z:/cases/case-a" will be accessible for both machines (same with the evidence data). Option 2: use network names This option relies on the fact that you can safely access resources on other machines (within the same network) using their network paths. So you could tell Connect to share a case from folder "\\Processor-One\cases\case-a". Our goal is to wrap this up with a feature called "shared folders" that will allow to define those in the admin UI as well. Such shared folders will then be rendered in the folder trees under a dedicated branch.

Hi Adam! There are few rules that govern how quick tags are populated. I am listing them below: most recently used tags are prepended to the list. If you tag with tags A, B and then with C then the list will show tags: C (most recently used), B, A. you can also PIN tags to the certain slot. You do that by right clicking on the Tags Facet in the main Connect screen. If tag is pinned then it will never go off the list and it will remain in the fixed position. You can also make use of ALT+number keyboard shortcut to quickly assign a tag to the current item (so ALT+2 will apply the tag which is at the slot number two) Quick Tags are stored in browser's local storage, meaning that they are sensitive to clearing operations. You could try manually edit this piece of data, but I wouldn't advice it unless someone accepts the risk. Yesterday I fixed an issue with deleted tags still appearing in the Quick Tags list. If you are interested in getting access to this build, please open a support ticket and let us know which version of Connect are you currently using.

Hi Paolo! I'm afraid current technical implications make this task very hard to accomplish. We surely want this feature, but I can't promise if and when this will be delivered. It's not possible at the moment, but we will add such a feature in the near future.

Why not simply re-use export templates? I can imagine the benefit of exporting multiple formats at once, but I can't see myself going through an export wizard that includes combined options for PST and Load Files exports. Also with Connect your exports happen in the background. Therefore you can quickly fire two different exports (based on templates) for the same data sets and let them finish later while you focus on your review.

Hi Adam, we had this task planned internally for quite some time. I assigned this a 1.9.1 priority, so chances are you'll see that soon (can't make any hard promises as for the deadline, though).

Adam, I'll talk about this "tag locking" idea with my colleagues. Seems like a good concept, so I'm wondering if ever considered it and if other customers have asked for it before. The 2nd idea about rudimentary search UI has been already asked few times, but not recently. In general I'm ok with such feature, but it seems a lot of work. So I think it would be better if we gradually add more permissions that can hide certain parts of the UI until case managers can customize such interface themselves. We have few of those on our radar already (like hiding Cluster Map) so it might be simply a matter of time.

Thanks Adam for posting your findings here! Others might find themselves in a similar situation.

Hi Adam! To quote the User Manual "When multiple includes are used, the results are filtered for all items that are in at least one of the include sets, i.e. it is like filtering with the union of all includes.". So in your case you will see all items for given email address that are either an email OR in the specified date range.

Hi dpmills! Migrating cases from 1.7.x to 1.8.x format is currently not officially supported. We don't have an official list of features that rely on 1.8.x case format either. Our recommendation is to recreate those cases in Intella 1.8.x. That will make sure that you are using the most stable and mature version of our software and your case format will be up-to-date with newer versions of Intella. Moving user defined annotations should be possible based on the MD5 facet and manually tagging items in new case (or by using "Cross-case work reports" if you own an Intella TEAM license). Some additional attention would be required to see if results based on hashes lists are correct.

Hi Questa! I'm happy to inform you that the next version of Intella (desktop line of products) will contain more advanced keywords statistics, so I'm hoping you will find there something for you too. At the moment we are not planning to make the input data configurable, but it's certainly a valuable addition and I think it will one day end up in Intella. We are focusing on keywords for now because that is something that has been asked for the most. As for your second question, all those needs (reindexing or deletion of a particular source) have already been recognized by us and we are finally at the stage where we can start making changes into the right direction. The first step that we took is to allow Intella to pick up new or modified sources from given folder and index just those. If everything goes as planned, then this will be available in 1.9 and more features will come soon afterwards. I think I'll leave this post here as it revolves around the topics that are already on our roadmap.

We already have plans for that - it waits for a free time slot You can make use of "Print Report" feature to get the listing of attachments in your printout. Print Tab (in Desktop) and Print (in Connect) are just a way to get nicer printouts of what's currently being seen on screen (current tab). Ok, I'll talk this though with the team and if they agree I'll pull it onto the roadmap. That's still on our radar, but not an easy task to implement right now. Interestingly, we are not receiving that many requests for that feature anymore.

I agree with Adam that Option 2 should give you what you need, Phil. As for the proposed "include email children" option - were you thinking about adding it to the keyword searches? It feels to me a bit more natural to do it in a standard way: first select some results, then apply "Show children" action and then further cull down your results.

@Jason, you are just right. The idea would be to select a subset of items and run pre-processing on that set. @Adam, I think we would add it as a part of our automated tasks framework. Seems like a best fit.

@bob.stanley, is that something that is a common feature in other eDiscovery software? The reason I ask is because it immediatelly rises questions like: what to do with items that have multiple keywords? does it help to know that item had keyword "X" without knowing where exactly it is? Or perhaps that is just for the sake of reporting/migrating/pre-processing the data?

Hi! If you can't make any sens of the log files, then please open a new ticket in our support portal and send those logs to us. I'm sure we will be able to help. BTW, did you have a look at both types of logs (server logs and case logs)? Thanks!

@Adam, I'm not seeing a problem here. Why not simply change Connect's port to any other value like: 80, 8080 or 10123?

I think that a complete solution should have a variety of a text analytic tools in it's toolbelt. Adam, for your example Concepts Search would be appopriate. You have a "Pyramid" concept which roughly splits into two smallers: "egypt" and "finance". This should be a part of unsupervised machine learning toolkit. Walt is also correct - if he looks for "fraud" he might also be interested in "embezzlement". And this is not something that can be easily inferred by machines themselves, therefore it feels more like a supervised learning (perhaps simple dictionary of synonyms would suffice, but machines can't create those themselves). This is a vast topic, but a very interesting one @Walt, having more examples or rough ideas will only make our job easier and increase the chance of having it delivered in Intella. So by all means please do share those!

Hi guys! Walt, to address you last three questions: Custodians support That's of course on our radar for quite some time. I'm curious to learn though, what sort of integration would you exactly expect? With Hierarchical tags support (should be out in few days), you will be able to easily create a parent tag "Custodians" and create few child tags ("Bill", "James") that would represent each individual. If you add two PSTs from James to your case then you can use Location Facet and assign "Custodians/James" to each item in corresponding locations. To deduplicate this set it's enough to query for the particular tag and use decuplication in Items Table to have unique results. I can understand that doing that for 20 tags (representing 20 custodians) can be a bit laborous, but wouldn't that satisfy your use case? Rest assured that we have custodian-level deduplication on our radar too, so sooner or later it will be added to Intella. I'd love to hear more ideas how to integrate custodians into Intella. What sort of stats/reports/queries support would you need? Simple Text Analytics All good ideas, I must admit. I'd love to see at least some of those included in one of the future versions of Intella, but it's not an easy task to do properly. I'll see if we already have some plans about it. One thing that I find as particularly useful is the idea of "Concept Search", where predominant "concepts" (terms representing some abstract ideas) are computed for each item and represented in a graph that shows relationships between them. How would that sort of tool fit into your needs? Two-factor authentication Another nice idea, but using some third party authenticator (like in this case) would require for Connect server being open to public. Not something every company would agree to. However making it an option sounds like a good idea, so I'll put it on our roadmap. I wonder if anyone else would make use of such feature?

I'll add it to the roadmap, let's see what we can do in a long run.

Yes, exactly right - it's applied globally. It seemed like a more flexible approach for us when we designed it.

Yes, that would be the most useful thing to do. However, it's also the most complex one :-) Keeping track of incremental changes applied to various databases (and we use quite a few) is a lot of work. Let me flip the question and ask - did you ever get into a situation where current backup procedure did not work or wasn't feasible anymore?