kurtosis-letter.pages.dev Open in urlscan Pro
2606:4700:310c::ac42:2d13  Public Scan

Form analysis
0 forms found in the DOM

Text Content

WHAT HAPPENED TO THE KURT APP

An Open Statement to the KU Leuven Community


TL;DR

 * We built Kurtosis, a faster alternative to KU Leuven's KURT library booking
   system, which was widely used and appreciated by students.
 * After spending 5 months developing a full mobile app version with enhanced
   features, they presented it to university management.
 * Despite initial positive reception, the university ultimately rejected the
   app citing security, privacy, and maintenance concerns.
 * We responded to these concerns with technical solutions but the university
   maintained their rejection.
 * We is sharing their story transparently and encouraging students who support
   the initiative to contact university leadership.


HELLO

As students of KU Leuven who have experienced the pain of booking seats with
KURT and saw other people struggling with it, we wanted to solve this problem
and share our solution with other students.

In early summer 2024 we publicly launched Kurtosis — a simple web page that
allowed you to see the availability of the all seats in any library with a
single click of a button.

The response was overwhelming. Post about Kurtosis on Reddit became one of the
top-10 upvoted posts of all time on r/kuleuven. It was used by hundreds of
students during summer blok, and people keep using it at the time of writing
this post.

We realized that we were not alone in our frustration with the official KURT
interface, and we put out a survey on the Kurtosis page, essentially trying to
figure out "Why are you here and not on the official KURT web app?" and also
started researching by talking to people in person. The feedback was clear: the
official web app was slow, clunky, and felt like filling out a tax return rather
than finding a place to study.


WHAT WE BUILT

Driven by the support from students, this motivated us to build a full mobile
app for KURT, that will be faster, more efficient, and user-friendly. The goal
was to build the product that we would want to use ourselves, and the focus was
on efficiency and empathy.

We assembled a team of three people who previously contributed to Kurtosis and
started by researching how students interact with library spaces and designing
an interface around it. Over five months since June 2024, our team developed a
mobile app for iOS and Android from scratch.


OVERVIEW

 * Isometric map of the library to see where exactly is your seat. Inspired by
   The Sims.
 * 4 times faster performance across all use cases, including search, booking
   and checking in.
 * Smart instant search called 'Auto' that finds you a seat even when the
   library is full.
 * Booking and checking-in with Face ID using phone hardware secure enclave,
   making it more secure than the browser. No KU Leuven Authenticator required,
   without compromising on security.
 * It uses existing KURT infrastructure more efficiently, reducing the load on
   KU Leuven server dramatically and requiring minimal ongoing support from the
   university team.


FIND YOUR SEAT ON THE MAP, NOT BY DECODING A SEAT NUMBER

Instead of endless lists and dropdowns, we created an isometric map view (just
like in The Sims), where you can actually see the library layout, the trees, the
windows, and most importantly – where you'll be studying. We wanted students to
feel like they're choosing a real space, not just clicking a database entry.


FIND SEATS INSTANTLY, EVEN WHEN LIBRARY IS FULL

One of the most impressive features we've developed is called “Auto”. It's an
intelligent alternative to random search, enabling you to find study seats even
when the library is full and there are no seats in KURT.

Simply select the time frame you're interested in, and the app will locate a
seat if one is available for the entire duration, prioritizing your favorited
seats. If no suitable seats are found, the app will create a sequence of
available seats, optimizing for the least amount of seat switching.

Booking a seat or a sequence of seats requires a single tap. No need for going
through pages and pages on KURT to manually assemble a sequence of seats.


NO MORE EMPTY, BUT BOOKED SEATS

We know that many people take breaks during their study sessions, but the
current design of KURT makes it very hard to do so, especially during exam
season. And we don't blame them.

However, This leads to empty, but booked seats, and is frustrating for everyone
— both people who need a seat and people who took a break because they can't
make the seat available.

Our app doesn't limit you to select a single time frame for your study session.
You can add breaks between your study sessions, and the app will automatically
figure out which bookings to make. And again, it's just a single tap.


COLLABORATION WITH KURT

In September 2024, we had a meeting with the KURT team where we presented our
work. They were enthusiastic about the project and asked us to prepare a shorter
pitch for higher management. This too was well-received, eventually inviting us
to give a presentation at the Campus Bibliotheek Raad Humane Wetenschappen
meeting in October.

On October 21, we attended the meeting and presented our app, as well as
explained our plan ensuring sustainability and addressing any concerns regarding
support. When we finished the presentation with "And now we're ready to answer
any questions", we received no questions about the app at all, which we thought
was a good sign.

However, a few days later we suddenly received a rejection. It came in a short
email that started with "It was fun having you" followed by "significant
concerns" that have appeared. Specifically, they mentioned four points:

 1. Risks regarding security and privacy are "particularly high"
 2. Concerns about the lack of audit and monitoring capabilities
 3. Collaboration would require significant ongoing time investment
 4. Added value of the app is "limited" due to the map being in the roadmap of
    KURTv3, described it as "the major shortcoming of KURT today"

Following our presentation where no concerns were raised during the Q&A session,
we received a rejection email citing several concerns. Their email ended with
"We have decided not moving forward with this collaboration" and no further
actionable feedback.

After working on the app for five months, we were extremely disappointed by the
rejection, especially considering this important context:

REGARDING SECURITY AND AUDIT

 * Our app implements stronger security measures than the browser-based
   solution, utilizing phone hardware secure enclave chips. It's the same
   technology used by banking apps.
 * All data processing occurs locally on the user's device. It's never sent to
   our servers or stored anywhere other than KURT themselves.
 * We have been open to any possible security and privacy audits by the
   university or any third-party experts to ensure the app meets all necessary
   standards.
 * But even not considering these points, at the time of the rejection, we have
   been proactively reporting several major security vulnerabilities in the
   current KURT system.

REGARDING TIME INVESTMENT

 * The app was specifically designed to rely on existing infrastructure, it was
   a conscious decision to minimize the ongoing support required from the KURT
   team.
 * During development, we implemented authentication through an alternate method
   we identified in the KURT system, which bypassed the KU Leuven Authenticator.
   Our only technical requirement was official authorization access to KURT.
   Once granted, this integration would operate independently without requiring
   ongoing support from the KURT team.
 * Our team has also proposed a plan for a student organization to take over the
   app's maintenance and development, ensuring that the workload will be
   distributed among students and not fall on the KURT team.

REGARDING ADDED VALUE

While the KURT team has mentioned to us that some of the features we have
implemented are planned for KURTv3, such as floor plans, we still believe that
our app has a lot to offer, for example:

 * It offers at least 400% faster performance
 * Reduces server load dramatically, which was acknowledged by the KURT team. By
   our estimates, in some cases up to 350% less compute 1
 * Authentication with Face ID or other built-in biometry instead of KU Leuven
   Authenticator, without compromising security
 * Secure detection of users leaving study spaces with prompts to check out,
   eliminating the problem of booked empty seats in libraries
 * Provides unique mobile features not possible in a web app, such as Live
   Activities on iPhone


FOLLOW-UP

On the same day, we sent a response to the KURT team, asking for clarification
on the rejection. After we sent a response to their concerns in the rejection
and requested further clarification, the university's reasoning shifted:

 1. Claimed that they can't trust "any code that uses student data ... without
    full assurance of its integrity and purpose"
 2. Justified the approval of Quivr by claiming it operates in "read-only" mode
 3. They "appreciate proactive approach" in reporting vulnerabilities
 4. Ended the email by stating their decision derives from a trust issue, not
    "with you personally but with the potential risks involved" (whatever it
    means)

Therefore, since the university's decision to reject the app was based on a lack
of trust and not on technical or security concerns, we have no way to address
this issue. We have proposed solutions for every concern raised, but the
university has not provided any actionable feedback or a path to approval.

For context, other approved third-party applications like Quivr actively process
and store student data for features such as friend system or schedule sharing.
Meanwhile, our app - which only processes data locally and communicates
exclusively with KURT - was deemed a privacy and audit risk.


OUR POSITION

While we disagree with the reasons behind this decision, we are ready to
implement any changes and have proposed solutions for all highlighted problems.
As of now, we do not see a reliable path to our app's approval, so we are
disappointed to say that it is highly unlikely that our app ever gets released.

Our team has proposed solutions addressing all concerns, ranging from technical
integration to long-term maintenance. Beyond offering to implement necessary
modifications in collaboration with university technical teams, we developed a
plan for a student organization to ensure the app's sustainability. This project
started because we believed in KU Leuven's commitment of supporting student
initiatives and we hoped to contribute to the university community.

We're sharing this story because we believe in transparency, and because we hope
it might spark a broader discussion about how student initiatives are evaluated.
The code we wrote could have made hundreds of students' lives easier. Instead,
it will remain unused.

Below you can find a copy of all our presentations (both for KURT and for
internal meetings) ever made for the KURT mobile app project, as well as the
video materials.

 * Presentation 1
 * Presentation 2
 * Video 1
 * Video 2

We remain committed to finding ways to help fellow students and hope that future
initiatives will find more support in their efforts to improve university
systems.


WHAT'S NEXT

The KURT development team stated that the KURTv2 will be disabled soon, which
means that Kurtosis will also stop working as it depends on the KURTv2
infrastructure, and we can do nothing about it.

We know many of you have shared our frustration with the official KURT web
application — the performance, the interface, and overall user experience.
That's why so many students have embraced the Kurtosis app as a better
alternative. The positive feedback and growing waitlist make it clear there is
strong demand for a better solution.


WHAT YOU CAN DO

Despite demonstrating that our app is significantly faster and more efficient
than the current KURT system, the university has prohibited us from launching
it. This decision feels like a major missed opportunity, both for students and
for the university itself.

If you share our disappointment, we encourage you to make your voice heard.
Reach out to the KURT support team, or any other KU Leuven staff members you
feel may be able to reconsider this rejection.

 * Campusbibliotheekraad Humane Wetenschappen
 * ask.kurt@kuleuven.be — KURT Team
 * luc.sels@kuleuven.be — Rector of KU Leuven

If you believe this app would benefit the student community, please share your
thoughts about how a faster, truly mobile-friendly booking system would improve
your study space reservation experience.

If enough students voice their support, the university may be compelled to
reconsider this decision.

Thank you for your support — it means everything to us.

Sincerely,
Ilia, George, and Alexander

--------------------------------------------------------------------------------

FOOTNOTES

 1. Since we don't have access to the KURT server logs, we don't have technical
    capacity to measure the exact performance improvement. Therefore, we
    estimated the performance improvement based on the number of requests made
    to the server and the complexity of the requests.
    
    Example of the compute required for booking a seat with a one-hour break:
    
    * Data lookup call = 1 compute unit
    * Data write call = 1 compute unit
    * Seat search call = 4 compute units (due to the complexity of the query
      running on the database server)
    
    For KURTv3: 1 (library data lookup) + 4 (search for booking before lunch) +
    1 (book seat) + 4 (search for booking after lunch) + 1 (book seat) = 11
    
    For the app: 1 (bownload availability data) + 2 (book two seats)
    
    The same calculation was done for other use cases, such as booking or
    searching for a seat, and the results for some use cases were up to 3,5x
    performance improvement. While we understand that this is a rough estimate,
    when we presented this to the KURT team, they acknowledged that the app is a
    lot more efficient than the current system.