PleaseTech blog

We aim to provide useful, pertinent and sometimes fun insights into the world of document collaboration and the workings of a technology company

Every little thing they do is magic

Posted by Sarah Edmonds on 14. August 2014 09:51

The other half of marketing... Google


It’s our job in our marketing to translate the magic.  Over in techie heaven (as they’re fondly known in PleaseTech), once the team have delivered the latest product release, it’s up to us to communicate the upgrades to the end user.  Sometimes it’s a tough gig, sitting in a meeting trying to understand what they mean when they talk about database transactions, continuous integration, encoding or regular expression… However, those are just some of the tools that have been used to design PleaseReview v5.1, but what do they mean and how do they benefit our customers?

A database transaction makes sure everything or nothing happens in a transaction.  So, if you spend $100 on groceries, a database transaction makes sure the money is both debited from your account and credited to the store’s account.  When thinking about PleaseReview it keeps the integrity of the data in sync, so if you and e.g. Tim are both online at the same time reviewing a proposal, and Tim then makes and immediately withdraws a comment, you aren’t able to reply to the comment.  Sounds obvious, but if you could reply to a comment you’d briefly seen, that had then been withdrawn there could be lots of random responses applied to a document.

Continuous integration does what it says on the tin.  It’s a development technique which continuously merges to our development servers the work developers individually do on new roll outs and integrations.  Its main aim is to prevent integration problems and to avoid one developer's work in progress breaking another developer's efforts, thus allowing our teams in the UK and Malaysia to work more effectively together.  For our customers, it means a higher quality product with fewer bugs.  So for example, PleaseReview v5.1 provides ‘post review reporting’, which delivers metrics around a review such as ‘how many proposed or rejected changes were there?’  This ‘review data’ is delivered via an Excel spreadsheet. Another 5.1 enhancement called “sub-reviews” allows a reviewer on one review to branch off their own sub-review with their own set of reviewers and then merge consolidated comments back into the original “master” review. These things were being developed by different teams at the same time but, rather than only bringing them together when they are both complete, the process of continuous integration means that every day we can test the latest sub-review code with the latest post-review reporting code to make sure they work properly together.

Character encoding represents a repertoire of characters , which is used in both computation, data storage, and the transmission of textual data At PleaseTech we use a universal coding (UTF-8) that handles letters from all alphabets.  Lots of our customers require documents to be reviewed not only by many people internal to an organisation, but also people externally who may work in different countries.  Encoding means that as a document passes between computers in different countries whose first languages may be different (French, Spanish etc.), the document doesn’t become corrupted.

What regular expression means to me, and what it means to our developers are two different things… To our techies, regular expression is a sequence of characters that forms a (potentially complex) search pattern.  This supports the new context-based review feature of PleaseReview v5.1, which allows reviewers to search for a word or phrase within a document to ensure that e.g. lower or upper case is being used correctly, or that a word or phrase is being used in its correct context.

Of course, there are lots other new benefits that can be found within PleaseReview v5.1, which is rolling out as we speak, and was talked about by Dave earlier in the year right here on this blog.  To find out more or to experience a little bit of the magic for yourself, please get in touch.  

Introducing PleaseReview v5.1

Posted by David Cornwell on 2. April 2014 09:34

Founder/CEO of PleaseTech Ltd - collaborative document review and co-authoring for the enterprise.


April already. They say that ‘time flies when you're having fun’. I can tell you that it certainly flies when you are trying to get a software release out.

The testing of PleaseReview v5.1 is now well under way and, assuming no major issues are identified, we expect a release date towards the end of May. We never release until we are sure it’s a quality product and our testing is complete. There is testing of the new functionality, regression testing and installation and upgrade testing and, of course, all documentation and other support material to prepare. There are so many elements in the mix.

So what will PleaseReview v5.1 contain? Well, as always, the thinking behind the release is to:

·         Continue the ‘beyond review’ strategy;

·         Facilitate enterprise rollout with enterprise enhancements;

·         Address new client requirements;

·         Keep up with the changing environment.

The ‘beyond review’ strategy’s intention is to consolidate our thought and technology leadership by adding value to the review process.

So, what does PleaseReview v5.1 include?

Sub-review and parallel reviews

One of the new features in v5.1 will be the concept of a sub-review. This will allow a review participant to create a sub-review, review the document(s) with their own chosen review participants and then publish selected comments and changes from the sub-review to the master review. To explain:

Imagine I was a department head and was invited to review a corporate policy or procedure which affected my department. I would want to first discuss this with my management team to get their feedback and consolidate their comments. Sub-reviews are designed for exactly this scenario. We can have an ‘internal’ departmental review and then publish our consolidated feedback to the master review without having to ‘wash our dirty linen in public’.

Parallel reviews are somewhat different. This would be appropriate if you wanted to gather feedback from two entirely separate groups at the same time without each of the other groups being aware of the other’s existence.

This increases the ‘workflows’ available in PleaseReview so with v5.1 there will be, out-of-the-box:

1.       Standard single stage collaborative review

2.       Sequential reviews where each stage can comprise one or many participants;

3.       Sub-reviews;

4.       Parallel reviews

Combinations are possible. For example, if permitted, it will possible for a participant in a stage of a sequential review to create a sub-review. This allows for very sophisticated review management.

Context-based review

Most people see review as an ability to comment upon and mark-up a document. And, whilst that is correct, there are many ways to look at the document. You can sit down and read it as you would a book. You can follow all the cross references and therefore jump backwards and forwards in the document. Or you can ask the question: ‘What does the document say about nnnnnn’?

It is this latter approach which context-based review is designed to support. Reviewers are able to search the document for a phrase or text string and PleaseReview will produce a report with all instances of the phrase (or text string)  presented in context. This is much like a Concordance, whereby a list of the material words used in a work are presented together with their immediate context as a separate index.

Providing the context in the report is important. This allows the reviewer to rapidly scan the report and examine the document for consistency. It’s more than just checking for spelling or capitalization. It allows the reviewer to check that a phrase or term is used in a consistent way throughout the document.

These subtle requirements come from being involved in endless discussions in respect of document review and from listening to people struggling with these issues. By listening to our target audience and then incorporating their requests and requirements into our product plans, PleaseReview continues to set the standard for document review.

Post review reporting

Post review reporting will further extend PleaseReview's ability to deliver metrics around the review. Whilst with the current release a set of comprehensive review metrics is already available, these are mainly delivered at a document or system level. For example, how many reviewers made how many comments and what percentage were accepted or rejected, etc.

The post review reporting available with v5.1 will allow companies to drill down deeper within the document itself.  So, for example, let’s examine section 3 of the document. How many accepted proposed changes were there? In section 4 of the document how many rejected proposed changes were there?

This requirement has been driven by one of our clients who is looking to use review metrics to analyse the quality of writing and reviewing. By examining the number of accepted and rejected changes on different sections of the document, some initial determination can be made of the quality of the author and/or the reviewer. At the very least, flags can be raised as to which areas merit further investigation.

This illustrates that reporting can bring real value by helping to control and measure the review process. In all honesty, it's not the primary reason customers turn to PleaseReview but is simply a welcome side benefit.

Enhanced configurability

As PleaseReview gets rolled out across large organizations, the requirements of many thousands of users need to be addressed. For example, the comment categories which may be appropriate for an engineering document will not be appropriate for those in marketing and will be different again to those required when reviewing a proposal.

We could have done the basic minimum but we took the plunge and have implemented a full hierarchical inheritance model. What this means in English is that we will deliver a highly configurable system which retains central control with the absolute minimum work required. It is possible to specify the behavior of the system with respect to a specific department and/or review type (for review types see below) and have one override another. So, for example, if a review type permits the download of the original document and the departmental settings do not, the departmental settings will override the review type. We believe that this level of configuration will serve to meet the requirements of large enterprises going forward.

Cost Center licensing

An aligned but separate requirement is that of cost center licensing.

A large company may have a single installation of PleaseReview, but licenses are purchased from individual departmental budgets. These departments may take a dim view of another department using licenses purchased from their budget.

Cost center licensing will allow groups of licenses to be ring fenced for an individual department’s use.  Once again this facilitates enterprise deployment and, hopefully, keeps peace and harmony in the corporation.

Review types

Facilitated by the enhanced configurability, review types take PleaseReview's current templating capability to a whole new level.

Now standard review types can be set up which specify all review parameters including, potentially, the duration of the review, the review participants (via standard distribution lists) and a host of other configuration parameters.

So, for example, as someone who has just written a blog entry and wants it reviewed, there could be a standard review type called ‘Blog entry’. So all I have to do is upload the document, select the review type and the review will be started for a set duration to a standard set of people included on the ‘blog review’ distribution list. If, in the future I need to change who reviewed the blog entry, I wouldn’t change the review type I’d simply amend the distribution list associated with the review type 'Blog entry'.

Of course, for larger companies, it would be possible to have subsets, such as engineering blog, marketing blog, etc.  

This is especially powerful when coupled with standard workflow systems such as those found in a document management system (DMS). The DMS user simply initiates a pre-set workflow which in turn calls the PleaseReview review type. In this way a sophisticated, integrated system requires very little work.

Archive 

Finally, we will be offering an optional archive module as a cost option.  When we initially conceived PleaseReview we saw reviews as transient instances which, when the document was approved, would be discarded. The approved document would be the true electronic record and how it got there was immaterial.

However, as we have started branching out into new market sectors and people are placing a greater importance on due diligence, compliance and being able to prove that company procedures were observed, several clients have requested the ability to archive review data.

The archiving module will meet the needs of these clients by ensuring that the review data is securely archived prior to a review being remvoed from the system.

Other stuff

Additionally there will be support for new environments. As a company offering on premise solutions (in addition to the cloud), we operate in a complex, ever changing environment as other companies upgrade their offerings.

Needless to say, with a release of this magnitude there are always minor enhancements and bug fixes included. These are too numerous to mention.

The only constant is change.

I’m confident that with PleaseReview v5.1 we maintain the high standards PleaseReview has been setting for years and that PleaseReview will continue to lead the market with respect to document review.

header bg