Some inofficial hints how to succeed with the main administrative tasks of your friendly ecole doctoral of IP Paris

ED IP Paris - Admin Stuff

Disclaimer: On this page, I collect some hints how to successfully complete some of the main administrative steps of doing a PhD in the ecole doctoral (ED) of IP Paris. Needless to say, only the official reglement interieur (RI) of the ED, hopefully easily findable on their webpages, is the only binding text. Still, reading and following the hints below might ease your life. My last version of the RI is this one.

This page is surely not yet perfect. If you find a mistake, a part that is hard to understand, or you feel that something critical is missing (that is, simply reading the RI did not resolve your problem), please contact me (Benjamin Doerr).

General rule for PhD students: I am happy to help you, but this is not my main job (which is to be a professor at Polytechnique). So your first point of contact is your supervisor. If this fails, don't be shy to contact me, having your supervisor in cc. It is important that your supervisor is informed as (s)he might have a different view on things. Since it is hard to educate you to stick to this rule, I will ignore any emails not having the supervisor in cc. Of course, if you want to discuss some matter confidentially, then you don't need to put your supervisor in cc.

General rule for all: My role it so overlook the scientific side of your PhD projects. I can do nothing on the administrative side. Here you should contact the right person in the doctoral school administration, which almost always is Audrey Lemarechal, in particular for "I need to register/reinscribe by day J, but I get this document only by day J+x" or "Can I replace the insurance certificate by ..." or "I entered nonsense in ADUM and do not know how to correct it". Emmanuel Fullenwarth is the person to contact for anything related to defenses. It is a good idea to put me in cc when you contact the administration.

Reinscription (updated 07/10/24)

Each academic year anew, you have to reinscribe by paying some fees, filling out some forms, and declaring on ADUM that you want to reinscribe into the next year. The details of this process change every year, and usually without that this is announced. So keep an eye open on what is actually required, talk to other PhD students (ideally in the same year), and check this page where I try to list the most common problems. If you see that some data on this page is not accurate, please let me know.

Here is a list of hints in random order:

Please do all steps of the reinscription in close interaction with your supervisor. There will be questions you cannot understand, and if you answer at random, you'll run into trouble. If your supervisor has no time for that, let me know and I'll find a way to reduce his/her supervision load to a level that allows him/her to do his/her job correctly :-)
Please double-check that everything is correct before "finalizing" your inscription request. You cannot change the request after you have declared it final other than by re-starting it after I rejected it or begging Audrey Lemarechal to undo your finalization. Either implies that all "avis favorable" have to be redone.
Please follow all advice you find in ADUM, on the webpage of the ED (note in particular the "useful documements" tab), or which you received from the administration via other channels. I don't give a complete description here, but only try to add some clarification to points that are often misunderstood.
With your reinscription, you declare that the data in your ADUM profile is correct. Hence please check that it still is. You might have moved, your PhD topic might have changed mildly, the supervisor team might have changed or their involvement rates might ask for a modification. All this can now easily be changed, but is hard to change outside the reinscription (you need to beg Audrey Lemarechal to do that after getting my OK, a process that in total can take some time). Needless to say, all changes to the PhD project have to be discussed with your supervisor. If you have a very active supervisor (in the sense that (s)he has an official supervisor role in many PhD projects), let him/her double-check that he/she is still in the limits of what is allowed (6 students as thesis director and 300% supervision rate if no derogation was obtained beforehand).
Start the process in good time. You need to collect many documents, some of which might be time-consuming (e.g., it takes a few days until your payment is taken into account and you can download the certificate). You might make a mistake and need to restart. Your inscription request needs to be validated by your supervisor, the head of the lab, me, the head of the ecole doctoral, and the ED administration. The deadlines change every year, but for 2024 you have to finalize your complete and perfect inscription request before October 20th and this needs to take all further steps before October 31st.
On document you need is the report of the annual CSI meeting. Consequently, you need to (i) declare you CSI to be if you're inscribing into the second year and you haven't done so (see below for more details on the CSI), (ii) organize the CSI meeting (with report, presentation, discussions), (iii) get a report on this meeting, (iv) get your supervisors to sign it, and (v) finally upload it to ADUM with the reinscription. The current announcements state a deadline of October 20 for this, but since October 20 is also the deadline for the whole inscription, you better do that earlier.
If you would like to inscribe into the 4th or higher year, you need to ask for permission to do so by adding to your inscription the derogation form (found on the ED webpage) filled out and with all signatures. Good news: If you defend your thesis before the end of December, you don't need a reinscription. Bad news: Every academic year such that you have been a PhD student for at least one day in that year counts as year. So if you start your PhD in June, certain administrative steps come a good deal earlier than if you start in October. These is no bonus for this, sorry. Of course, your reinscription into the 4th year will be granted for sure (but you have to fill out the form at least in some minimalistic manner, ideally clearly mentioning what will be the total duration of your PhD).
Usually I suggest that you send me an email if you need me to validate something on ADUM. This is not needed if you reinscribe before the deadline since I know that there's work to be done. If you don't manage to reinscribe before the deadline: Try to get it done as quick as possible. No need to warn me since there is nothing I can do. Most likely, the ED will wait for some time before they complain. Notify me when the dossier is ready, that is, you finalized your inscription request and your supervisor has validated it.

Comite de Suivi Individuel (CSI, english: monitoring committee, updated 03/09/24)

The precise nature and use of the CSI is described in Section 3.3.1 of the RI. Please read this short paragraph to understand the main points. The following two paragraphs discuss two questions that are not completely explained in the RI.

When to Declare a CSI?

You have to declare the CSI at least before the reinscription into the second year, simply because you need the CSI for the CSI meeting which you need for the reinscription. There is no need to wait until then, you can to it anytime. After six months is good moment. Note that the CSI can be helpful also beyond sitting in the annual CSI meeting.

Selecting a Good CSI

Before thinking about how to fulfill some formal requirements, it is a good idea to take a moment and think what the CSI is good for and what it can do for you (it sometimes gets forgotten, but most of the rules of the ED are made to help the PhD student).

The aim of the CSI is to have additional experts that overlook the PhD project (including the first steps after the defense) in some mild manner. They can give additional ideas or point out aspects which the supervisors or the student might have not seen. They can be approached by the student and by the supervisors at any point of time, and this without that a clear need is apparent, when a problem is about to show up, or when there are real difficulties. I note that, in principle, with such questions you could approach any member of the LIX and they will usually be happy to help. Experience shows that if there is someone who’s official role is to help, approaching him/her becomes easier.

When selecting a CSI, in addition to the formal requirements, I would consider the following points.

Ideally, the CSI members are not too close to anyone in the PhD project (as this might make it hard to give neutral advice). In any case, for both supervisors and PhD student there should be a CSI member to whom they would feel confident to talk in case of a real conflict (happens very rarely, but if, then the CSI comes in handy).
Giving advice is easier if you have already seen a lot of life - this is why the RI asks for experienced researchers.
The CSI is not part of the supervising team (for this, there are the supervisors). So it is less important that they are the utter expert for the topic of the PhD project. Rather, it is important that they have a good general understanding of what it means to do/supervise a PhD in France and what is going on in areas close, but outside the core area of the PhD project.
NEW SINCE 2024: Members of the CSI can be members of the PhD jury, but THEY CANNOT ANYMORE BE RAPPORTEURS.
Strategic thinking (this is not my own wisdom and I’m not sure how relevant it is): The members of the CSI are somewhat forced to follow your works and progress. This can be used as marketing instrument, that is, you try to have people in the CSI whom you want to impress with your work.

The formal rules are: Both are not part of the supervising team (in the broadest sense). At least one needs to have the HDR (or equivalent) at the moment of declaration. At least one needs to be a specialist in the area. At least one needs to be external (not IP Paris, HEC, and whatever we join with in the future; not working in the Turing building; a part-time teaching job at Polytechnique is OK).

How to Declare a CSI?

To declare a CSI, please follow these simple steps.

As laid out in the RI, the CSI is fixed by the ED, following a proposal by the DT in agreement with the student. Stick to the formal requirements (3.3.1) and use common sense (it is possible to choose a committee that fulfills the rules, but not the purpose, and this will not be validated).
The DT proposes the CSI to me by filling out this simple form and sending it to me (with the student and all other supervisors in cc). Wait for me to tell you that the committee is good.
Bring your CSI into ADUM. This will not be validated even if ADUM suggests that there exists some validation process.

Rapporteurs and the Jury

As always, let’s first understand what these are good for, because then it should be quite obvious what to do (namely not go for the weakest legal solution).

Good Rapporteurs and Juries

The rapporteurs and the jury assert that your PhD studies are worth the desired diploma. They do so formally and informally. The formal defense process ends, hopefully, with the doctoral degree. Nevertheless, in particular if you’re staying in the academic world, there is more written (hidden) on your degree than the binary yes/no information. For good reason, the rapporteurs and jury members appear on the title page of your thesis and in many other official documents. They give credibility to your degree (same as the institution, by the way). This credibility is a soft criterion. If you have a PhD degree, then it’s a valid degree even if your jury is crab. But a hiring committee might ask (in their internal discussions) why you have such a weak jury, and they might suspect that the quality of your PhD was not so great. So in your own interest, I’d suggest that you go for great rapporteurs and juries.

So what makes a credible jury? Of course, people

that understand the research done in the PhD,
that understand what a PhD means,
that are impartial,
that are in a position to possibly speak up if something is wrong,

AND where this is visible to the outside (via publications in the same area, via successful PhD supervisions, via a perceived distance from the PhD project, by being at least twice as cool as your supervisor as perceived by status, publications, citations, etc.). It is very hard to find people that tick all the boxes. Someone understanding very well what you do is very likely to be, at least somewhat, a friend of your supervisor. For that reason, the jury can be up to eight members, so that you can split the competence over different people. Note that it is not wrong if some jury members are not exactly from the small field of study of your PhD - you want someone to certify that this is a "PhD in computer science" and not just a PhD in "theory of ant colony optimizers".

Formal Rules

The formal rules try to enforce that your rapporteurs and jury have at least some minimum credibility. The details are laid out in the reglement interieur (RI), the only binding text (together with some hard-to-find juridical texts). Right now (2024), the rule are as follows.

Rapporteurs: At least two rapporteurs, but two is really the standard. Both (i) need to have the HDR, be full professors ("professeur") or directeurs de recherche, or count for some other reason as HDR-equivent, (ii) be "external" (not involved in the PhD supervision and from outside from IP Paris and HEC and any other institution taking part in the PhD supervision), (iii) cannot have any publication with the PhD student or a publication with the supervisors in the last three years (note: every publication counts, even if you didn't ever talk to the person or if it is just a tutorial or ...), (iv) and cannot be members of the CSI (since 2024).
Jury: The jury consists of four to eight members. The thesis director is necessarily part of the jury, for the rapporteurs it is common that they are members of the jury, but it is not required. (i) At least half of the jury has to be "rank A", that is, full professors, directeurs de recherche, or equivalent (see below). (ii) At least half of the jury has to be external. (iii) If the jury has at most five members, only one supervisor can be part of it (necessarily the thesis director), otherwise at most two supervisors are admitted. The jury shall contain a fair number of women, where fair means similar to the gender ratio in the scientific community the thesis belongs to. CSI members can be part of the jury.
President: The jury selects one of its members as president of the jury. The president must be rank A and cannot be a supervisor. It is very uncommon that the president is a rapporteur and it should be avoided. It is very uncommon that the president is internal (IPP, HEC). The president should be present onsite for the defense (in case of doubt, please double-check the momentary rules with Emmanuel Fullenwarth). Needless to say, any jury must contain at least one member who could take the role of a president.
Invited participants: There is the official role of the invited participant of the jury. These are people you want to have in the defense and participate in the discussion of the jury, but who for whatever reason cannot be jury members. A very typical case are supervisors who cannot be in the jury because the number of supervisors is limited. Invited participant can talk, but not vote (same as the thesis director).

These rules are obviously not complete (e.g., they contain no clause requiring any expertise). However, "La composition du jury est validee par l'ED, suite a son examen par le responsable de domaine, qui peut demander des modifications" will ensure that there is no problem. Note also that the rapporteurs and juries are also validated by the president of IP Paris and there have been cases where juries were rejected for being to light-weight.

Particular questions: For foreigners, there is the questions if they can count as having an HDR or even as "professeurs ou assimiles" (which means "rank A", that is, PU or DR). There is no complete answer to this, but this conversion rule is a good guideline. Things are even harder for industry people. Since a PhD is an academic degree, here I tend to be strict. This does not mean that I disencourage having industry people in the jury, absolutely not, but I usually find it hard to see them as HDR-equivalent or even rank A.

Emeriti: The jury may contain at most one member with an emeritus status, this person can even be a rapporteur, but not the president of the jury. A number of additional constraints have to be satisfied, see the RI.

Declaring Rapporteurs and Jury

The formal procedure is that you put your rapporteurs and jury (with some details) into ADUM at least two months before the defense (together with a provisional date), I validate them, and some levels above me validate them. If you follow this road, you do everything correct. If you also follow these two suggestions, you make your life easier.

First, as for any other validation, ADUM does not notify me when something needs to be validated. So I suggest that you send me a short email.

Second, if I cannot validate your jury, then you cannot just fix this. Instead, you have to run to Audrey Lemarechal and ask her to re-open the form for you. For that reason, I suggest that you first send me your rapporteurs/jury so that I can check if everything looks good. Please add affiliation, precise job title (e.g., MdC, associate professor, Hochschuldozent, ...) and whether you see them as rank A or B, HDR-status, has at least somewhat co-supervised or not, the gender you will put in the ADUM form, link to homepage, link to Google scholar profile, and possibly further data that might be useful (e.g., a CV for industry people without a detailed homepage). For the rapporteurs, please state what is the most recent publication with any member of the supervision team and with the student. Please confirm that you checked that there is no conflict of interest (if in doubt, describe what could be critical).